The web began with server-side rendering, and with some extra help from jQuery, we were happy.

Then came the early 2010's, and a new generation of framework emerged, lead by AngularJS, and ushered in the client-side rendering revolution.

In the last 3-4 years, there's been another revolution, lead by ReactJS, but this one is silent - hybrid rendering.

In this talk I'll go through the benefits and pitfalls of both "classic" rendering techniques in historical context and nowadays, and showcase tools such as NextJS and Gatsby and how they allow any developer to improve their product in ways that seemed impossible just a few years ago.

With decentralized applications we can build unstoppable apps with no single point of failure.

BitTorrent was one of the most popular dapps ever created. At one point it was estimated that BitTorrent traffic accounted for over 50% of total internet traffic.

In this talk we will learn about the ingenious p2p file transfer protocol by implementing a bit torrent client. We will discuss the finer points of the protocol and understand how the BitTorrent protocol can be used to reduce the server and network impact of distributing large files, reduce dependence on the original distributor and much more.

Audience members will walk away with

• Understanding the underlying concepts of p2p file sharing.
• Understand the pros and cons of p2p protocols and file distribution.

In this talk, we'll build a gatekeeper to your website. Our fraud detection system will target various types of malicious activities, such as account takeover, parameter tampering, forbidden access and more. We'll try to identify potential attacks and react to them in near real-time. Addressing this problem in a classical batch fashion will result with a complex, non-scalable nor real-time solution. We'll adjust our clumsy implementation to a modern, stream processing windowed-aggregation, use Kafka Streams as our streaming framework, and end up with a beautiful, clean and maintainable code.

Writing user input validations is something most front end developers get to deal with during their careers. For some reason, this simple enough task often adds complexity and bloat into the developer's code.

We borrowed concepts and syntax from the world of JS unit tests to create an open source framework that handles the most common pains of user input validations.

It's going to be a surprise this year, it's going to be.... interesting... ;)

Bash is a great scripting language, available almost everywhere, quite easy to use and very powerful. Bashir is an opinionated development framework for Bash, written in Bash. The goal of Bashir is to ease the development process of tools written in bash, providing dependency management, code reuse, and some other common functionalities.

The bad kind of code reuse, with everyday examples. I will present (in a humorist way) the reasons for avoiding code reuse. Also, by attending, you might learn a new cake recipe ! I will also cover a few practical DO and DO NOT advices for real life coding. Be sure to bring your "ctrl-c" and "ctrl-v" buttons along.

When learning from complex data such as medical scans or video feed we encounter long training periods and large memory demands. This makes searching for the best set of hyperparameters a hard to nearly impossible task. Most researchers simply try a few sets of hyperparameters by trial and error or use a grid search at best. These hyperparameters are usually chosen by some intelligent guess based on prior knowledge. It is clear to see this is not the best practice. A more DATA DRIVEN approach required.

In this session, we will learn practical methods to search for the best set of hyperparameters under an arbitrary limit of resources (run time or machine cost) by selecting the correct optimization algorithm and training settings.

Embeddings in general and Word embeddings, in particular, are real-valued entity representations able to capture context semantics and trained on unique corpora. Models proposing these representations have gained popularity in recent years, but the issue of the most adequate evaluation method still remains open. I will present an extensive overview of the field of words and non-NLP embeddings evaluation, highlighting main problems and proposing a typology of approaches to evaluation.

BPF (Berkley Packet Filter) has many uses and is used extensively in the Linux world. It is used for observability, for controling packet capture with Wireshark, and for data processing in XDP. The cillium.io Linux uses BPF exclusively for network configuration.

However, using BPF once one thing and developing a portable, predictable BPF program that works with different kernel versions is an different story. Getting BPF to work is made even more challenging since many important details such as the ELF file format and kernel error messages are scarcely documented. In this talk, I'll take you through the process of creating a stable BPF program that grabs interesting bits out of TCP connections at high throughput. I'll discuss what you should do to get BPF to work, what you shouldn't.

As data scientists in healthcare, we wanted to enhance our experience in Deep Learning. Fortunately, the "Machine & Deep learning Israel" community managers decided to frontal teach, for the second time, the Stanford University course: "Convolutional Neural Networks for Visual Recognition". We decided to jump at the opportunity.

In this lecture we will present our course final project, answering the following question: "Does Transfer Learning using a network trained to classify age and gender improves the detection of medical findings in chest X-ray images?".

This project is a collaboration with Zebra Medical Vision and is supervised by Ayelet Akselrod-Ballin, Eyal Toledano and Eli Goz. Extra credit to Amir Bar. Our classmates in the project are Netta Shemesh and Gil Baron

What was once a bad dream is now a reality: any developer can push any code to production at any time (and sometimes multiple times a day) without any sign-off. In order to cope with this fundamental change, we need to surrender control and put more trust in our developers. But how? How can you build a decentralized security approach? What tools are available? How should your work culture change to support this new reality? In this session, you'll hear about the journey we've gone through at Soluto – what has worked, what hasn't, and what we're still working on. You’ll gain some practical advice and learn about tools that allow you to "let go" without compromising on security.

Software world is changing rapidly, and software developers in organizations are finding it difficult to keep up, and have no time to learn new technologies, although they are probably used by other groups in the organization. In the long run, this can affect developers’ satisfaction and their employability.

Adopting Inner Source (the act of leveraging Open Source Software development practices within the confines of a corporate environment) seems like the answer.

Elbit's Inner Source program relies on the Software Guild (an innovative gamification approach) and the Skill-POOL - a market place encouraging developers to experiment with new technologies as a meaningful on the job training.

The talk offers an overview of the journey of the Inner Source program.

The developer to manager path is a two way road. We are Rona and Rona, working in the same team, both on the road, but in opposite directions. One transitioning from development to management, and one from management back to the IDE. And surprisingly, we’re both going uphill.

When is the right time to take these turns, and why? What are the risks?

In this talk we’ll each give her perspective, talk about how can one role benefit from the other, and share from our experience.

Diversity is more than just a buzz word– it’s a proven way to generate innovation, increase productivity, and meet business goals. However, as the talent demand continues to grow for hi-tech R&D teams, the massive gender gap among them still remains. Currently, women only make up 23% of Israel’s tech professionals and only 16% of its tech management.

So...why should you care?

Tapping into the knowledge our team at Joonko has gained over the last few years, this session will prove that diversifying your staff does not have to be as hard you think. In fact, diversifying your R&D team can lead to a big ROI. Learn how to uncover the hidden biases that are holding your team back and build a more inclusive, innovative workforce.

We read other people’s code 80% of the time. While our code feels like a backyard playground, others code tend to be a scary jungle. Why is it like that? And what can we do about it? Let me show you some eye opening tips for turning this frustrating day to day task into a nice - even preferred - one (hint: it can tune your brain, just like training a machine learning model!)

We will present our journey from batch-based to real-time analytics. We implemented a data-pipeline lambda architecture using Spark Streaming for real-time analytics. We will introduce the main components of this architecture and will overview its different use-cases. The first use-case is to build and run real-time predictive analytics using Contextual Multi-Armed Bandit models for UI optimisation. The second use case is to do CTR (click through rate) estimation based on real-time data using weighted linear regression models.

In the end of the session, you will learn how to use Spark to combine real-time and batch analytics, and become more familiar with Spark's capabilities.

Companies struggle to keep the balance between happy engineering to advancing business features. Most fail at it, and accumulate large technical debt creating frustrated engineering. This can be done differently when combining Technical Leadership from the CEO combined with a unique strategy for combating Technical Debt. In this lecture I'll explain how the strategy of building a Technical Roadmap composed of Goals and milestones for each, and interweaving it into your ongoing planning cadence, helps you change the way you view Technical Debt and effectively eliminate this concept all together.

Static file hosting - from your CV to large media files - is a handy tool for sharing files that are either too large or change frequently. URL shortening enables you to track the usage of links you share online, and make professional-looking links that start with your own domain name. In this session I will show how I created these two services with Serverless tools with a near-zero cost and ~200 Lines of code. We will review the design and dive into code. This session should be helpful for developers who wish to build Serverless applications and reason about their architecture. No prior Serverless knowledge is assumed.

Hiring? So is everyone else. How are you going to make sure you get the best minds on your team? You train them! Yup, your read that right, Mentorship is making a comeback.

While experienced workers may be easier to onboard fast, in today's startup economy there are clear benefits to hiring less experienced (potentially qualified) people who are just getting started. Mentorship has its challenges, but Junior Techies are more available, cost less and don't bring their wrong old habits to work.

A mentorship program can be painlessly applied with a few simple tricks, to help your understaffed team as well.

LOL, I hope you didn’t take the title too seriously… Two things are sure in life: death; and the fact that no one anomaly detection algorithm can rule them all. Anomaly detection has various benefits for almost any company - system and business wise - but to make such a system valuable, there are multiple challenges, ranging from algorithm theory to system operations you should be aware of.

In this talk we will go through those challenges and the theory behind that field, and how you can get valuable insights.

API is the best way of connecting software engineers and sharing data, so building the perfect API for your partners can be quite a challenge. Partners can differ vastly in their requirements, programming languages and even API semantics. In this session, you will learn how to build a delightful API experience for your partners. I'll talk about delivering a clean API, easily generating documentation, mock creation for testing, cross-platform client generation, all within a predictable release cycle and without extensive effort (I promise :) ).

You've been hired to migrate an organization from Ruby to Scala—two very different languages. It's your first day on the job. Now what?

In two years Riskified went from 2 Scala developers to 30, over 7 training rounds. Starting from a narrow focus on the immediate future, our mindset gradually shifted to "how to write good code" to "how we code".

In this talk I will describe how we evolved from training developers to forming a community inside the company, and share with you some strategies appropriate for growing companies.

The next major release of Angular will bring to the front stage their 3rd and most promising renderer engine, IVY. What's making IVY so special? What causes Google to invest so much effort in re-writing their core framework renderer? What's the difference between the last view engine and IVY? What is incremental DOM how IVY will make all FE development better and faster? Join me to discover anything you need to know about IVY and Angular upcoming release.

Let’s face it, recruitment is hard. We want only the best engineers to join our incredible team.

We, as interviewers, have a leading role in the recruitment process as we have to assess whether a candidate is a good fit for our team. We sometimes forget that we have significant impact on the interview’s atmosphere, which can greatly affect the candidate’s performance.

This short lecture will pinpoint some of the things that you, as an interviewer, can do to make the interviewing process a lot nicer for candidates (and yourself). Ultimately, helping them be at their best and therefore, making the decision a lot easier.

Now more than ever, illustration is an integral part of the product experience. Illustrations can be found in many areas of a product – guiding users through onboarding, helping them understand how to perform complex tasks, and shaping their emotional response.

Whether you’re working on a B2C product for the family or a B2B product for a large enterprise, using the right illustrations can help create easier-to-use products and facilitate more impactful relationships with your users.

In this talk, I’ll show you how to decide where illustration is crucial in your product experience – and where it isn’t.

As Israeli developers, we want software to understand Hebrew text. Natural language processing (NLP) for Hebrew is necessary for creating AI applications. As of 2019, this is far from trivial:

1. The computational problem of understanding Hebrew is hard.
2. Hebrew is a “small” language, with lower commercial appeal.
3. Frameworks and tools for NLP in Hebrew are far and few between.

In this presentation we will explore the need for open source Hebrew NLP tools; the problems involved; which tools are available and what we want to improve, hopefully with your help. We believe that closing the gap in this field requires a coordinated effort, led by Academia and supported by government, volunteers and hopefully the industry to create the necessary open source Hebrew NLP tools.

So, you have an open source project that's successful. Awesome!

You have millions of downloads, things are looking bright and it looks like everyone wants to get a piece of you. Just what you're you were hoping for.

Slowly, people start making demands about your project. You start feeling pressured. Working on the project isn't so fun anymore and you wish all the pressure would just stop. What happened?

In this talk I'll share my story and the story of other maintainers maintaining several open source libraries with 5M+ downloads a week.

I'll share my failures in keeping people I love engaged as well as my successes after learning what helps.

Startups often outsource their development, but how often are they really happy with the results? Outsourcing is often considered a "magic solution", making the contractor the magician - but even the best magician needs the right settings to work their magic.

In this talk we'll discuss the most common friction points I've encountered while working with various contractors, and how to minimize (or even avoid them altogether), for better outsourced results from day one.

Apache Spark and Kafka are essential tools in Big Data technologies. Apache Druid is a high performance analytics data store for event-driven data. On my talk I'm going to present how we used these technologies for building realtime analytics on top of billions of served recommendations. I'm going to describe the problems that we encountered and lessons that we've learned.

Money, discount, debt, interest, regulation, incentives, and auctions are terms taken from the field of economic and are relevant for software development.

In this talk, I will give examples of how I used economics concepts and apply them to software development. This how I managed to take better decisions, motivate my team, and to improve code quality,

Ever thought about presenting your work to others? Talking in a meetup or a conference? In the past I couldn’t even think about it, I thought that it’s not for me and I won’t get any benefit from it at all. In the last year and a half, things have started to change. I was always curious and looking for ways to improve as an engineer, so I did what most of us do: read tutorials, posts and watch technical videos. Still, since I came from a different background from most engineers, I felt like I had a knowledge gap that blocked me in many ways. The journey that I will describe in this talk will show how I improved my technical and soft skills by helping others in platforms like stackoverflow, writing blog posts and talking in meetups.

All programs have errors. Some are caused by the environment, some by user errors and some are due to bugs in your code. Yes, even yours. The success of your application depends on how you handle those errors, so you better have a good strategy in place. In this talk I will discuss different approaches to error handling, their advantages and disadvantages, and how to choose between them.

Requests per second, request latency, and overall system performance and reliability are some fundamental concepts that need to be taken into account when designing a high capacity API. With CI/CD becoming a common deployment methodology, deployments to production are a constant occurrence. How can we ensure that the core capabilities we initially designed our system with to have remained intact? We built Predator. It allows us to manage the entire lifecycle of stress-testing a server, from creating a test file, running scheduled and on-demand tests, and viewing test results. Its seamless integration with both NOSQL and SQL databases keep test results persistent. Bootstrapped with a user-friendly UI alongside a REST API, Predator helps developers simplify the performance testing regime.

This is a story of a race against time! So hang on to your seats...

During a customer migration to a new attribution system, a huge project for AppsFlyer Engineering in 2018, we found ourselves facing a potential data loss catastrophe. It all started with the primal sin of a premature optimization made where we set the incorrect data retention timeframe for a database holding 65 billion records.

When we discovered this, with only one week to respond before the data is permanently erased, we channeled our Macgyver skills and got to work. During this session I'll describe the chain of events that brought us to the cliff's edge, the steps we took around the clock to save our data, and how we managed to forestall any data loss for our clients.

This talk is not about micro-frontends.

At Trax, we use multiple frameworks to build a unified applications suite. The challenge of creating one visual language led us to a journey to find the most abstract way to implement reusable components.

In a midst of a polyglot of frameworks, a ray of sanity and abstraction shined to guide our way out of the abyss. In other words, Web Components are awesome.

In this presentation I'm going to tell you our story, the technologies we investigated, the components we threw to the side of the road, and how data-binding and style-encapsulation can be made beautifully regardless of the wrapping framework.

After working in Scrum for years and struggling with its pitfalls, we've decided at Tipalti it's time to try a different project management approach. We've started a journey wherein the end of it the entire organization was working in Kanban and practicing Continous Delivery. During this journey, we've learned many lessons about Kanban and about transitioning to it from Scrum. In this talk, I'll share the most important ones with you.

Chatbot interactions are becoming the next thing in development cycles. One of the areas that can benefit from this is the CI/CD—the idea of a simple: Start interactive session with the CI/CD infrastructure instructing it to build artifacts, spawn and tear down environments, deploy, rerun tests — everything you need to get to the finish line. Our journey started when we realized we want a to move from traditional CI environments to an on-demand, short living CI environment. We insisted on making it a developer-driven experience.

The secret sauce is combining Jenkins-X’s serverless version, GitOps and some other goodies. We’ll be happy to demonstrate how to put them together In the end, we realized that this also cuts the costs for cloud infrastructure. Let’s change the way we brew CI/CD.

Companies no. 1 threat for growth is their challenge to operate at full scale due of their inability to hire great talent, or hire almost great talent fast enough. This trend is expected to get worse before it gets any better and so, we are faced with two options: Hire tier 3-4 engineers or rethink the problem as a whole. The new workforce generation is moving towards more non-traditional employment- freelancing There is a huge pool of highly talented freelancers, who you are not interacting with today. It is time for organizations to start experimenting with on-demand talent pools, build knowledge & methodologies, create practices and muscle to operate at full capacity In this session we will review the landscape, opportunities and look at some examples that are thriving in this economy

How do you drive your people’s personal commitment? If you look at management from the point of view of leading volunteers, how to make things happen without the power of “real authority” and increase the employees’ motivation and ownership. As a manager of both volunteers and employees, I’ll share my insights about dealing with those challenges. I think that the approaches for the two might be much closer than one might think.

A bot that sings to you when you're sad? Generative poetry derived from classic literature? Paintings that draw a composite of every person that's ever looked at them? The cutting-edge web technologies that can power these creative applications are easier to use than ever!

Modern JavaScript conventions, browser APIs, AI services and IoT hardware provide a unique platform for artistic expression, exploration and creation. In this talk we'll examine this type of interactive application development and look at the open-source projects, APIs, services, hardware and approaches to creating expressive works through JavaScript — audio, visual and otherwise.

Data is one if not the most important component today to make a great machine learning process. but data itself sometimes isnt enough and you have to process it in some way to make it ready for machine learning, this process can take time. In this lecture I will demonstrate how serverless computing can provide a great benefit for AI data preprocessing. as an example I will take face recognition deep learning using the Watson Machine Learning service, while letting PyWren with Cloud Functions do the data preparation phase. This makes an entire process up to 50 times faster compared to running the same code without using serverless computing.

What exactly is DevOps, and what does a DevOps do?

In this talk, I will focus on the main role of DevOps as I see it: eliminating fear!

Devs have plenty to worry about: How do I deploy to production? How do know my code isn’t going to break production? when production break, what will I do?

Management also has its share of anxieties: After a production issue - how do we make sure it won’t happen again?

DevOps have one core fear: How do I know everything is working?

We’ll discuss some best practices: testing, CI/CD, monitoring tools, and postmortem.

But with a twist: using these tools not just to solve technical issues, but as a way of creating a cultural shift within the company.

The end-goal of our talk is simple: Learn how to move fast and break free of your fears!

Israel has many social NGOs that help lot of people. Most of them are technical clueless and don't have anyone that know how software can help them. They desperately need help in defining and writing software tools (like mobile applications, web application etc).

Pro-Bono of software engineers is great help for such NOGs.

In the short speech I will demonstrate what we do in MicroFocus to help NGOs and what we learned from over the last 6 years.

It is not uncommon to hear about how elegant Functional Programming is. Granted, some notions of beauty do have their place in the world of software development, but that alone is not a sufficiently compelling reason to endeavour into the unknown.

In this talk, I will describe my (somewhat idealized) journey from the world of imperative, Java-style programming down the never-ending path of Functional Programming. We will be motivated by simple and relatable concerns, such as: the wish to have less bugs, to write simpler tests and to have altogether more predictable software. This will almost inevitably lead us to re-discover and embrace Functional techniques and the benefits that follow.

Tracking users is not only IP recording. You can use a lot of tools to track and analyze users across platforms even without IP. In this session, I will explain how users try to conceal their identities and how to bypass those techniques. How you can detect users and track them even if the use incognito mode. How to track them even if they use VPN or TOR. There are a lot of open source tools and code that is available to help us: from digital fingerprint to forever cookies. From WebRTC local IP detection to DNS cookies. You can use the information in this session to violate's users privacy, or you can use the session information to know how to defend your privacy. I use it for the latter.

Last September I participated in HackZurich, Europe’s largest Hackathon. Hundreds of engineers from all around the world gathered for 40 hours of coding, pizza, and NO sleep. The Hackathon offered 16 different coding challenges. We chose to tackle the “Millenials on the move” challenge, provided by Esri and Zurich Insurance. The mission was clear - to use geospatial data to save lives on the roads. In this talk, I will show how we came with an innovative approach to solve the challenge, which led to the winning solution. We came home tired, happy and with awesome prizes!

"Why Two-Thirds of the Fastest-Growing Companies Fail?" is an article by the author of "Startup Leadership". His answer to the question was "these failures can be avoided by consciously creating a culture that simultaneously values the competing ideals of innovation and efficiency"

My story begins 1.5 years ago in Taboola Engineering. I found an organization that grows FAST and the need to learn and share your knowledge grows accordingly.

By the end of this talk you will know how to improve people’s performance by creating a culture of sharing. If I had to summarize it in one sentence - find the people that shine, use their passion for making your company grow.

Let's create a game set in space!

For this, I have to render a nice looking galaxy. How do I do it? Do I create a pre-rendered image and place 100k stars manually (oh no!), or do I try to create something random procedurally that looks awesome?

In this talk, I'll delve into the basic concepts of procedural generation - mostly around the basic mathematical tools at your disposal as a developer to create awesome procedural generated art.

With every step in creating a realistic looking galaxy, I will provide code examples in JavaScript.

SOA has been around for decades, and its latest iteration - microservices - for a while now. Just five years ago microservices were hip, dominating the agenda at conferences; now we almost take them for granted. With microservice-focused conference talks losing steam, the time is ripe to consider what we, as an industry, can learn from the microservice disruption; more importantly, we have a golden opportunity to consider what we have yet to learn!

In this talk I'll analyse the lessons hammered into us as we've adopted microservices, and take a hard look at the challenges I believe we still struggle with as an industry. If you're not doing microservices you'll leave the room knowing where the pains will come from; if you are, you will leave with ideas and leads on mitigating these pains.

The home assignment for tech interviews became widespread. Companies like them, as they save interviewer's time and present an objective scale. Interviewees should like them - They work on them when it's convenient, not in the middle of the working day, without standing nervously in front of the whiteboard.

If you're a junior developer, this is your chance to demonstrate how you're better than your years of experience. Prove that you know the best practices and principles of building a small project, instead of just talking about them - Show, don't tell.

Join and learn what the company actually looks for.

Navigating the pitfalls of the home assignment will land you more options, better offers than before, and open doors previously closed. And eventually $$$ in your pocket.

The whole point of running a data lake is to let users consume the data and make powerful business insights. But sometimes, as expected when you have many users, things will go wrong. And when they will - you should be ready.

In my talk, I'll present an incident that happened to us a few months ago and how monitoring the usage of our users (with the ELK stack) helped identifying the source of the problem.

I'll also explain why it took us so long to find what caused the problem, and the practical conclusions we made.

There’s a lot of hype about the risks of using open source components. (http://techgenix.com/linux-security-concerns/) Whether it’s a highly publicized court case or a huge data breach. After 10 years as a developer and one year as the head of an R&D group in an open source security startup, I learned first hand that a lot of the risks are manageable. In this tech session, I will demo open source vulnerability, and assess how a vulnerability in one of the most popular open source libraries impacts your code (or not). This session will provide developers with a deeper understanding of how open source vulnerabilities affect their code, and the practical tools to manage and mitigate their open source risks.

You had a great Idea, you wrote the perfect program for it and it's working. Now you have a few customers and they are thrilled with your product. So you think to yourself what's next? why not sell it to everyone? Oh wait, you can't.. your system will crash under the load. Recently Check Point managed to overcome the challenge of scaling up its system to support bigger customers, and now there is nothing we can't do. In this session I will describe the difficulties we faced and how we managed to handle them:

• Monitoring: Identify bottlenecks in the system
• Scale In:
  • Save time with bulk operations.
  • Add optimizations based on system behavior.
• Scale Out: Make adjustments to enable auto scale.
• Deployment: Migration and scale up was done with zero downtime and no customer impact.

How, when and where did the podcast and the conference came to be. Stories behind the scenes and the unwritten (unspoken) history of technical content in Israel’s tech

AI became the biggest buzzword in the tech world for the past few months. Somehow – it always feels like some rocket science reserved for data scientists or python developers far away from the front end. In this talk, Eliran will show how easy it is to create your first prediction model with a simple web application using TensorFlow.js and other deep learning tools that can run on your browser! No python, no server, no data scientists – just javascript.

Streaming algorithms are everywhere, and are relevant for any developer that has to deal with scale - whether it's message streams or queries on big data.

While many developers will just treat them as black magic, a slightly more in-depth knowledge of them can be interesting and critically useful.

In this talk, I intend to:

• Explain what streaming algorithms are,
• Show how, in Glassbox, we used them in original ways to implement otherwise impossible solutions,
• Present the difficulties and surprises,
• Present a small catalog of useful algorithms that you could find original uses for.

After this talk, you will have powerful new tools in your coding arsenal.

There are several practical tips that can make interaction with Stack Overflow much more pleasant.

In this talk we'll go through the common reasons people have a bad experience with Stack Overflow, how to avoid them and how to ask questions like a pro in Stack Overflow.

Hopefully, I will be able to save people the pains I had to go through when answering questions in Stack Overflow.

On the threshold of the twenties, we rave about serverless computing and artificial intelligence, autonomous cars and the Internet of things. And at the same time, we continue to bump into the same walls when it comes to collaboration between humans. But there are also models that work! Let's talk honestly: why is it so hard for us to work together and does DevOps help with this? And if not - what can help us?

The rise of serverless computing platforms is a major force that creates new opportunities (and, of course, risks). In this talk we will focus on the LANDIS stack which emerges as a dominant pattern for quickly delivering applications and services that are inexpensive but, at the same time, are inherently scalable and reliable.

Specifically, LANDIS stands for: Lambda API Gateway Node DynamoDB Identity (Pool) Static (web-site hosting)

Discussing the various components of this stack, we will cover patterns, practices and pitfalls. For instance, we will explore topics such as: is API Gateway always needed; why is Node the recommended runtime for the backend (Lambda); how to protect yourself from financial exhaustion; codebase structure; and, the elephant in the room: cold starts.

When asked to create an FaaS solution I was confused. What is serverless? What is FaaS? Why are we here? It seemed like everyone had a different answer.

In the end, I realized I had to create my own personal definition: • Allow the end user to post a snippet of code.
 • Run it on demand.

Two steps, that’s it! 
Learn how you can do it in 3 bash commands in under five minutes with Knative!!

As Wix Backend grew exponentially with more than 800 micro-services, our previous CI solution has been slowing us down. We decided to switch to the Bazel build tool and Google Cloud Build CI platform. The migration and adaptation was challenging and some promising directions had to be abandoned. The end result was a dramatic improvement in performance and accuracy.

What makes bazel stand out as such a fast build tool? Is the migration to it hard? What are the pain points? Come and find out from our experience.

Software-Engineering and Animal-Rights are two passions of mine. When I do one and not the other - I feel something is missing in my life. About 3 years ago I've started my journey as part-time programmer, part-time Activist (also a programmer) - and I'm not going back!

In my short talk I would discuss the challenges and advantages of this life-style for the programmer as well as for the employer, share some funny moments, and talk about the challenges.

I would like to encourage the audience to take that path - to make their lives - and the world - a better place.

Building a large web application usually requires a state management library. up until now Redux and Mobx were the two major players, but with the new Context api and React Hooks, it seems that there can be an alternative for developers looking to reduce the boilerplate code while maintaining clear state separation

Predictive innovation is something humanity has always been struggling with.
 We, as product builders, lean too much on gut feeling and a bit of luck when launching new products or introducing new features.

 People don't just buy products; they hire them to perform a certain job.

 In this session, I will introduce Jobs-To-Be-Done theory that helps us understand the causality behind people's choices and deconstruct their decision-making process so we can create a replicable process to uncover people's unmet needs and build better products for them

What is software architecture? This seemingly simple question has baffled researchers and practitioners for decades. Industry legends such as Mary Shaw, Martin Fowler, Grady Booch and Ralph Johnson have weighed in on just what architecture is; but as an industry, we have yet to come up with a compelling description or concrete requirements from the role.

In this talk we'll consider software architecture in a new light, centered around one key observation: good architecture isn't about dealing with current requirements, but rather about avoiding future constraints. With that understood, we can have useful discussions about what is or isn't good architecture; hopefully, you will leave this session with ideas on how to build/evolve your software towards an unknown future comfortably.

When I started my army service in Mamram, I saw for the first time a giant mainframe computer - the IBM 370. It contained 1.6 MB of memory, divided into 8 slots of 200KB, one program runs in each one. That was 40 years ago. Since then the world of programming has changed dramatically, but some lessons can be learnt from the old days: can you imagine writing your program on a piece of paper, then printing it into a bunch of punched cards? Or that following an exception you'd get a hexadecimal memory dump, on paper (!), and this was your only debugging tool? Working in the industry today, I can sometimes hardly believe it myself. Software design was key at that time, dry runs and a lot of thought before actually running anything live. Is any of it still relevant today? Let's find out!

Your company grows and its business portfolio becomes richer. Multiple views are usually then split to different independent teams and each serve a different business functionality. Enterprise-grade software development will require that all of these applications will have a common product backbone: unified look-and-feel, single-sign on, shared entity management, access control, common services and so on.

Engineering organizations struggle to determine the right amount of decoupling, reuse, and interoperability needed to excel through this growth.

In this talk I will cover different approaches for building application suites and state pros and cons for each proposed strategy.

Like most startups we are running ab tests, like most startups we looked for an abtesting framework we can use and like most startups we ended up writing our own framework for abtesting. In this talk I will share our experience from a 3 years and hundreds of tests perspective a on the decision to write our own abtests framework. What worked for us, what did not worked and what you should consider when you find yourself in the same crossroad.

As we build products and features, one of our main goals is for our users to actually use them. One of the best ways to increase product adoption is to create Passive Impressive features - those that remove the need for the user to opt in without being intrusive or “Passive Aggressive”. Or in other words - opt out features that encourage users to ask “Why Shouldn’t I use this?” instead of “Why Should I?” In this session, I will talk about why we should focus on “opting out”, why it doesn’t apply to everything, and give tips on how to increase the focus on this in your organization.

Weighted lotteries are used when we need to randomly select k of n items with given probabilities - commonly used by Outbrain's recommendations system to provide diversity for our users. Once we identified that our naive implementation is a production bottleneck, we decided to solve the problem in a fun way, and turned it into a coding challenge, followed by releasing creative benchmarked solutions along with classic algorithms to the community

https://github.com/guyko/weighted-lottery/wiki

You fear your code. Once in production, you are afraid of changing it. A bug? You fix it using messy, yet surgical, hacks, rather than a refactoring that leaves your code better.

Admit it! You fear your code. But there's a way out: just write tests. Writing tests is impossible! It makes my productivity tank! My boss doesn’t leave me any time for writing tests!

These are all excuses for not writing tests. I will show you techniques that will force you to write those tests, and I will show you how they help you sleep better at night, make you more productive, and make you a much better developer.

Running a multi cloud deployment - something that quite a few organisations have pursued in the past few year, is a pretty big challenge. In this session we will talk about the motivation of running multi cloud, best practices, recommendations and lesson learned from our experience.

You have a great idea for an app or an online service. You're going to be super rich and conquer the world, all the girls would want you and all the boys would want to be you, it's going to be so awesome and mom's going to be so proud, unless... someone have already done that before you.

Let's face it - right now, while you're wasting time reading this blurb, it's either someone have already implemented your idea, or they are in the the final process of releasing it. That's why apps, and digital products in general, need to be built and out in the market fast. Like lightning fast.

"Micro Frontends" is the new buzzword in the Frontend world, but too many times people use it in the wrong context or with different things in mind. Micro Frontends can refer to different kinds of solutions that solve different types of problems - starting from using different UI frameworks on the same app or letting different teams work on separate parts of the code independently. In this session, we'll go over the different problems we have when building a big app and how different micro-frontends solutions can help with this.

Things that I've learned as being a tech entrepreneur and as a software engineer is that it's super important to know to say "No" to things, more specifically choose which things to decide not doing. It comes in mind even when trying to tell the difference between a Junior Engineer and as Senior one, the ability and the skill to know what NOT to do. Telling from experience of more than 13 years as a software engineer, more than 5 years of running developer communities and over 3 years as an entrepreneur, I'll mention all the mistakes a long the way, trust me, it's not a success story :)

In this talk we'll cover 4 families of programming languages you should learn.

We'll discuss how each of them will affect the way you think about and solve problem and will show some code examples from each language.

We'll explore some project idea for you to learn each of these language.

Bringing Data Science to Production poses some unique challenges, such as how to mobilize large code bases written in a variety of technologies and tools, some of which not optimized for production resiliency and scalability. In our talk, we will share Gong.io's experience, presenting our current Data Science CI / CD and production flows, what works well, and what we plan to improve.

Postmortem is never about pointing fingers, but what if it is just you doing the postmortem, and you're the only constant factor?

In my talk you will learn about my unsuccessful interviews, why I failed, how I learned from my mistakes, and how did I work on not repeating the same mistakes, to be more successful in future interviews.

Most importantly I will talk about why it is essential not to play the blame game, but to use postmortems as a tool to learn about yourself and how to improve.

Deep Learning is everywhere, it is the future, it overlooks the backend and the frontend. In this lecture we overview this field and provide a good playground to start from. I will give a good understanding of its jargon, we will highlight the difference between deep learning and machine learning. In addition, we will review the model types (CNN, RNN... etc’) and will touch on the differences between Supervised and Unsupervised. I will continue to talk about the tools that are common today and how as a developer you can harness them to help you. I will try to remove the barriers from this field.

There are a number of reasons for deploying a multicloud architecture, including reducing reliance on any single vendor, cost-efficiencies, increasing flexibility through choice, adherence to local policies that require certain data to be physically present within a specific area/country, latency reduction, and mitigating against disasters.

We spent two years designing, preparing, and implementing a full scale mutlicloud architecture. We invite you to learn from our journey which ended when our product - composed of 100+ services - seamlessly function on either AWS or GCP.

Questions, observation, assumptions. We keen to succeed and yet many of us fail to find the right formula. We utilize a set of primitives to accomplish that mission - asking questions, observing, and using our own assumptions to fill the voids.

But perhaps because we're asking the wrong questions and reaching the wrong conclusions during our cognition, we fail that mission without a clear understanding.

This talk will challenge the way we approach our learning & conclusion. I will try to answer how we can improve our chances to nail that interview, deliver that feature we're working on, or just to be a better professionals.

I will try to demonstrate with real world examples how by rethinking the question - by focusing on the why, and not the how, we could achieve much more.

For a while the ML engineering focus was around the Data Scientist needs, libs like scikit-learn, TenserFlow and more, serves that need. The industry started to change the focus and put it more on the post-DS phases, we can see big companies like Uber developing internal platforms like Michelangelo to address these needs, but not enough mature open sources are out there yet. That gave us @Fundbox the motivation to develop a ML Monitoring platform. In this talk I would like to walk you thru the product architecture, decisions we took on the way and the end solution. We will talk about the differences between stability and performance tests, what is a "concept drifting" and what kind of widgets we developed to reflect our models quality in production.

Once upon a time printing was a major part of any system developed. Documents, reports and manifests were the ultimate output. But today everyone hate paper. Not only it's bad for the environment, you can't even zoom-in or search in it!

Alas, sometimes a web project demands you to enable your users to print from the client side. So before you dive into the muddy pool of paper margins, page breakers and footer-header sections let me tell you about the print @media query and how it allows you to develop a print layout which is non intrusive to your on-screen design.

In this talk I'll share all the secrets, hacks and tips to make you a lean mean printing machine.

Starting as a new developer in a new company can be a stressful experience. In your first day you're expected to:

• setup your dev machine
• get credentials
• get the app working
• understand where the company KB is
• understand who can help you out

It is very time consuming (you always forget something), Not to mention the confusion of new recruits "what else do I need to do?" 🤷‍♀️

so how can we use tools that developers already feel comfortable with, to make the onboarding process easier and more fun?

Imagine you can run a script on your first day on the next day you can start to be part of the team (just like anybody else).

I'll try to present a simple concept of how to onboard with a one-stop shop script. that include provisioning, git, slack bots, docker and other automation.

Elasticsearch, the leading search engine and real-time analytics platform, can scale to hundreds of nodes and easily accommodate huge amounts of data while still preserving sub-second query latency. How cool would it be if we could run it on top of Kubernetes, the leading container orchestration platform?

As it turns out, we can do that easily and safely!

Join us in this talk where we will walk you through how to deploy a production-grade Elasticsearch on Kubernetes while adhering to best-practices of both technologies, keeping your data safe and your availability high.

We will also discuss scaling and auto-scaling of Elasticsearch and the Elastic Stack components while running on Kubernetes, and how the Elastic Stack components can be leveraged to a successful and happy deployment.

Looking at an everyday 3D printer - the printing procedure looks quite natural and straightforward. However, like traditional construction techniques things are not so simple. Indeed, just as scaffolding is extensively used in many building techniques, similar physical support is needed during the 3d printing process; different 3D technologies use distinct scaffolding methods. In this talk, after introducing the 3D printing ecosystem in general and the celebrated polyjet(™) technology in particular will describe the intrinsic difficulties of supporting the polyjet printing process, and hopefully demonstrate bone-fide algorithmic solutions to these problems.

It is a common illusion that Product Managers in the current reality, can continue working to define products, without thinking about security aspects, and that this can be done at a much later stage in the product lifecycle. How can we turn the Product Manager to be a real security ambassador? We’d like to show how a very comprehensive security program, managed by a dedicated security office in cooperation with a professional Product Manager, can raise the “security organizational conscience” and can successfully integrate security aspects in every product, with emphasis on agile development lifecycle! We will describe a process to bring security to the early stages of product lifecycle, starting from the planning stages ( aka “shift left”), which is impossible without full cooperation.

Malaria is one of the deadliest diseases, but we can defeat it. In fact, many countries already have - for example here, in Israel, decades ago. But it is not easy - operations must be planned and executed perfectly, and many countries fail.
In this session, I want to tell you how we tackle this problem. I will not dive into technical details. Instead, I want to share my hopes and inspiration. Currently, Zzapp is one of the leaders of Zanzibar's malaria elimination strategy and is one of the top ten teams in the XPRIZE AI competition.

Data scientists are adopting modern frameworks like Jupyter, TensorFlow and Dask, while data engineers mostly use legacy tools comprised of traditional databases, BI, data warehousing and Hadoop-powered data lakes. Can we really keep going with two different silos and disciplines? GPUs are widely used to accelerate machine learning, while new libraries like NVIDIA RAPIDS make it possible to process data frames in the CPU and accelerate packages like Pandas and Scikit Learn. However, processing data with multiple servers, or larger datasets is more challenging. In this session, we will demonstrate how to automate the deployment of big data tools side by side with a modern data science stack and GPUs. This solution means multiple jobs and interactive notebooks now run over the same cluster.

This is the story of how a startup deals with an 8 years old code spaghetti. It's not easy and it sure ain't pretty - but it works and we have our path clear before us. If you are interested in how a startup gradually consolidates its 5 code bases and enforces code standards that enable reusability, testing and good architecture on production code - this talk is for you.

While building a micro-service architecture one is most likely to encounter the same common issues across multiple scenarios. One of the greatest challenges is to identify these issues, or patterns thereof, in real-time and to provide the optimal solution for the actual problem. In this session I will share some of the ad-hoc fixes and workarounds we have encountered during these last few years, I will go over common mistakes and pitfalls, and discuss the strategies needed to take when starting to think about micro-services

Shortly after we began developing at Logz.io, we began to envision a development environment in which developers would be able to develop a feature, end-to-end, without relying on other teams. This vision was, and still is, based on a value of accountability that is part of our culture manifesto.

In this talk, I will unfold the journey we made as a company to the current state of things, where development is 100% self service -- from design to coding, testing, deploying, monitoring and debugging. This story includes Docker, Kubernetes, Apollo (our open source deployment app), Jenkins and processes and methodologies we adopted and enhanced (e.g. The Diamond Model for testing).

Everybody talks about cracking the coding interview. Dozens of articles and heated discussions on how to prepare, what to ask, and even how to handle the infamous brainteasers.

But let’s face it: the way we interview engineers is broken, and there is a very good reason for that - we are engineers! We know how to write code, and fix problems, a whole other set of skills than assessing other peoples' technical competence. In this talk, we will share six mistakes we were doing while recruiting engineers and share the path to master the skill nobody talk about - being a better interviewer.

One of the many challenges faced in building AI solutions for healthcare is gaining clinical buy-in, convincing doctors to trust the ‘blackbox’. For them, it is a questionable science which they need to understand before trusting AI as part of their work. To bridge this gap, we have decided to expose the internals of our core algorithm in a scientific publication, in Nature Medicine. I will discuss why choose this path, what alternatives were considered? What’s painful about exposing your IP to the world? How to bridge the worlds of academia and industry research, build the story, prepare the code and experiments, and what workarounds we found to address seemingly impossible technical requirements. Finally, it’s retrospect impact, and how others can benefit from this story.

We’ve all experienced it: you’re working on a task, adding some code, and then you need to store some sensitive configuration value. It could be an API key, client secret or an encryption key ― something that must be kept secret. And this is where things get messy. Kubernetes has a promise to simplify this process by using the native secret object, which, as the name implies, can be used to store secrets or sensitive configurations. Unfortunately, Kubernetes secrets are imperfect. But no need to worry ― there are solid tools for storing secrets securely on Kubernetes platform. Join me for this session to learn how to develop a Kubernetes application that can keep a secret ― for real.

One of the main principles of Lean Startup is about creating quick build-measure-learn loops. While this is very helpful in finding market fit for products, there are occasions when setting and meeting a concrete deadline is the priority of the org. Think about a product launch, that should be coordinated with Marketing, Sales, Training etc. In this talk you will learn about practical solutions for managing successful development teams that:

• Provide predictable deadlines and achieve them
• Use lean principles and iterate quickly
• Eliminate waste / buffers as much as possible We will use a real-life case study of how we do it at Next Insurance.

Hi, I’m Tsaela, I’m the lead singer in a rock band, and I’ve been developing code and managing teams for the past ten years. It never never ceases to amaze me how much the two loves of my life have in common. Albert Einstein once said that “The greatest scientists are artists as well.” Great developers can be compared to other kinds of artists who join together to bring their visions to life. Talented musicians share many of the traits that we look for in good coders. They are detail oriented and notorious perfectionists. They’re at their best when they’re analytical, logical, and methodical. Sound familiar? That’s probably because those are the same traits that our best developers have.

In this session I’ll share my experience of building an R&D team to perform like a rock band.

AppsFlyer’s mobile attribution and analysis platform is used by the biggest and most popular applications on Earth, generating a constant “storm” of 70B+ events (HTTP Requests) on our microservices, cloud based platform daily. In this talk, we will share how we built our microservices architecture with Kafka as its core piece. With continuous growth in 10X capacity - we needed to “learn on the job” how to improve our Kafka architecture by moving to the producer/owner cluster model, breaking up our massive monolith cluster to smaller and less fragile clusters with more resiliency and redundancy.In short this talk will discuss the steps we took to support our constant traffic growth.

Introducing a CI/CD paradigm, which my teams have been implementing and using for the past 3 years. Branch versions allows stakeholders to test and verify different code variants without affecting production. every code change creates a branch version which allows product managers, QA teams and different stakeholders to test and verify the changes, without merging the code to master. In this lecture I'll demonstrate a basic implementation and usage for Branch versions step by step, explaining through implementation every concept and tool that I use.

Microservices are awesome! Serverless is the future!! Who doesn’t want to join the new revolution as we dissect, shard and split our evil monoliths into small, simple, stupid, smart, self-deployable units? Yet, after we finish tearing up our backend, we’ll need to face some new challenges such as joining our scattered data model, securing our different endpoints, and of course - creating a simple, robust API for our consumers.

For the last five years, we’ve been using microservices heavily at Soluto and we’ve encountered every one of these challenges. In this session, I’ll dig into GraphQL, microservices, serverless, gateways and describe in details and code how GraphQL can serve as a uniformed layer across different services/functions/apps and provide the glue we so desperately need.

Did you ever get frustrated spelling a password over the phone? "*Capital* B, and then the *digit* 6, not s-i-x". After making sure the language is not set to Hebrew, capslock is off and your dad can tell a zero from an O, you give up before getting locked out. Now, imagine this is how internet is.

15% of the world’s population experience some form of disability, many of whom consume the web using assistive technologies that basically function as a middleman. Often, due to lacking implementation of websites, these technologies supply a partial, sometimes unusable, experience.

In this session I will give a glance at how web is perceived with assistive technologies, explain how they technically work, and share some practical tips to make your product accessible.

At JoyTunes we run a fairly unique startup-within-a-startup pod structure. Harnessing lessons learned from Spotify, Netflix and a few others, we built the entire company around independent, fast moving and self sustained pods.

In this talk we’ll use real life examples, including listing the pods we currently have, projects we’ve engaged in this manner and successes/failures. We'll briefly cover the reasoning and core principles along with the wonderful benefits and the not-quite-as-wonderful challenges, after doing this (successfully) for a little over a year.

We, at JoyTunes, ran hundreds of ab tests in the past three years. Some are simple (red/green buttons), some are complex (full pledge features that took weeks to implement). Over the years we developed some intuition on our users - what is likely to work and what is not. But every once in a while came a test that just blew our minds - tests results that were so surprising that no one expected them. In this talk I will share a few of those ab tests and their unexpected results.

Probability and statistics are fundamental fields in our modern world, yet they are not widely thought and often poorly understood and often counterintuitive. As a result, many misconceptions, erroneous beliefs and fallacies have taken root in common thought. This talk provides examples of such misconceptions and aims to show how we can better understand the world by building our statistical intuition.

In the early months of our venture, we wanted to validate our idea for a chatbot. Instead of developing chatbot capabilities in our MVP, we built a system for human operators. We learned a lot about our users, and pivoted twice in 5 months (eventually switching to a web-app solution). In this talk I'll explain how to get into the mindset and principles for creating such a flexible architecture.

Ever felt lost in an open space environment? Having a hard time concentrating with your co-workers warring over the AC? Creating an effective work space is one of the most important tasks in a company’s lifecycle. Riskified has fully embraced open-space culture from its get-go. In this talk I will share how we learnt to not only survive its annoyances, but even thrive in what it offers in productivity, work culture and social relations.

What would Node.js look like if it was written in 2019?

Deno is a new prototype project by the original creator of Node.js - Ryan Dahl.

It was created with lessons/mistakes learned from Node.js in mind. In this talk we’ll talk about the future of Deno, its philosophy and how to check it out.

Have you ever found yourself struggling to build an engineering organization that is quality-driven with consistently great results? When we analyzed why we didn’t feel our organization was performing at the level we had anticipated, we reverse engineered this to fundamental issues with our culture. Once we started working on this it had a ripple effect to our hiring process & then our onboarding process as well. This talk will dive into how we refactored our hiring & onboarding to set up new hires for success from day one. This ultimately delivered a well-oiled high performing engineering organization through a practically applicable methodology that is easily replicable. This not only enabled us to improve the quality of our hires, but also retain excellent talent in the long-term.

Although working with RxJs Observables simplifies complicated algorithms, the observable timing is not clear. Although the event loop cycle is described in detail in literature, it is not clear enough regarding observables. Is it a macrotask or microtask? Is it like setTimeout or is it like Promise? The answer is - it can be both and we can control it with RX Schedulers.

There is a best practice of writing unit tests during the development. Some experienced developers say that developing and writing unit tests together is faster than to develop without writing unit tests. When we tried to adopt this method of writing unit tests, we found that it costs more time with no benefit: The tests never failed due to a real bug.

In this talk, I will present when unit testing gets wrong and why. I will describe what method we used to write unit tests with a positive outcome, how it relates to DDD and clean architecture and how to apply it to systems with legacy code. By the end of the talk, you will have the strategy for unit testing against the living products that seek to continue growing, even though they are all built of complex legacy code.

Text2Gig is our contextual text similarity recommendation model. This is our new approach on Fiverr, free text search and similar item text-based recommendation. I’m going to talk about all the process from A to Z in creating this model. I will start with ideation of this model, continue with the development and the framework for this model and for dessert I will saw you some use cases of using the model and the results of testing the model on live users

In Outbrain we’ve noticed that big data processing is tricky, It’s a challenge to estimate delivery time, deliver high quality ETLs and maintain them over time. These challenges make life hard for all parties involved: engineers, data scientists, managers and business . Many of these difficulties may lie in our scarce usage of known software methodologies in the ETL domain, Methodologies such as good variable names, useful comments, testing and runtime alerts. We’ll compare these methodologies prevalence in Outbrains code base vs Hive jobs vs Spark jobs and see how we bring more of our software best practices to ETLs

The way that we consume content in the web has come a long way from the days of text blocks and images to AR/VR and live-streaming nowadays. Have you ever wondered how the magic happens? In this talk we will cover some of the defining moments in the evolution of content consumption in the web. Then we will show the ease of implementing rich content experiences in comparison to 5 or 10 years ago.

At Firefly.ai we use autoML to build models for users from various industries. These models can achieve high scores, but this usually comes at the price of complex models that are hard to interpret. Understanding how our models make their predictions is important for quality control, revealing patterns in the data, etc.

Interpretability of ML models is a relatively new field that is researched through a variety of aspects. In this session I will review how we applied techniques such as sensitivity analysis, PDP & ICE, and LIME for explaining the predictions of generic, black-box models at Firefly.ai. Using two real-world use cases from the banking and real-estate industries, I will describe the pros and cons of each method, differences between the methods, and production considerations.

Web SDKs need to provide a host of capabilities & are a contradiction in terms - on the one hand, they need to be "fully baked" & "closed" in order to provide a uniform API. On the other hand, they need to be flexible in order to support future development & a wide range of clients. While this can be achieved by "baking" a custom SDK per client - this is not very scalable (nor practically applicable with a business in exponential growth). In order to be able to deliver on the promise of modularity, we wanted to enable users to decide which capabilities they want to enable, without having to define this in advance. This talk will dive into the development methodology we used in-house to support this and eventually, how we serve multiple SDKs in a uniform manner to a diversity of clients.

After training a neural network to decide whether an image is of a dog, cat, fox, or boat, we can do various manipulations to try and understand how it works. One method is to take a single class and optimize an image that would maximize the network’s conviction of that single classification. For example, we could optimize to generate an image that isn’t just classified as "fox" at 80% confidence, it’s 99.9% fox. For some reason, the result is not foxy at all, it’s a big pile of random noise pixels. We’ll discuss why that is, how to fix it, and other techniques to improve our understanding of what the neural network is thinking when it’s classifying. Maybe we’ll find the foxiest fox after all.

There is a better way of working a PC - keeping both hands on the keyboard and leaving the mouse alone. It has a steep learning curve, nothing like using a mouse and a desktop, but it is quicker and more powerful. In the presentation I'll talk about the principle of keyboard-only input, the tools that make the keyboard and ASCII terminal as powerful AS any modern IDE.

Our talk walks you through the process of re-architecting huge expensive cloud systems to run leaner and cheaper (MUCH, MUCH cheaper). We’ll show you how by walking you through 2 of our projects that improved performance by 10x and drove costs down by 90% - saving $2,000,000/year.

You will hear about where to start, what to look for, what ideas to try and which ones to drop. We’ll show you how we’ve solved these issues on the cloud, using Serverless, EC2, FAAS and Spark clusters. You will see expensive code examples and The Fast and Furious code that makes it CHEAPER.

At the end of this talk, you too will have the recipe for re-architecting your systems to run FASTER, BETTER, CHEAPER.

GIS software in Israel has for longtime been a monopoly. In recent years there's a shift towards Open Source both around the world and in Israel. But in our case, the de-facto monopoly makes this change more meaningful - both for technological and financial reasons. This talk would go over the technology used to innovate in this market - from server side components to web and UI ones.

This is the easy-to-follow story of the algorithms and optimizations used to write the fuzzysearch Python library.

It starts with the classic linear-programming algorithm for calculating the Levenshtein distance between two strings. From there I'll describe, step by step, the development of the algorithms used to implement efficient fuzzy searching. Optimization is at the heart of this, since the goal is to find fuzzy matches efficiently.

I will describe varied optimization techniques, from low-level bit fiddling in C extensions written manually and with Cython, up to analyzing the problem domain and reducing the search space. There's a surprise twist at the end which had significant impact on the project and is a great take-away, but I won't spoil it here!

As many big companies and organizations in Israel - the army does its share of Open Source software usage. This short talk will cover its uniqueness and how does a closed network organization creates and releases Open Source. Also, Open Source is used to revolutionize the way it interacts with the IT industry.

I have been an architect for over 10 years. Many engineers that I met over the years don't know what is the meaning of this position, either because this role doesn't exist in the organization they work for, or because they don't have a close interface with an architect. I get questions like: How do I know if this role fits me? what I can do to improve my skills and get to that place? In this talk I'll share insights about the role and the answer I give to these frequently asked questions.

Our SDK is used by more than 130K+ applications and is what delivers all of our incoming traffic from the different platforms being used. Recently, we encountered a critical issue, that prevented any incoming traffic from Android devices to AppsFlyer.

As part of our anti-fraud solution, we use the Dexguard algorithm to prevent reverse engineering, where the large majority of our 130K apps use the same tool. However, due to the issue, some source files in the SDK were corrupted on the client-side, breaking the backbone of the SDK, basically shutting down the outgoing mobile SDK traffic to AppsFlyer. This talk will dive into the timeline of the incident, decisions we took in real time to mitigate the incident & the lessons we learned that we think can help others in a similar situation.

In our days, more and more products rely on 3rd party APIs to consume services and transfer data, in and out, while building their product value on top of it. This might sounds like an easy task, Rest/GraphQL/Messages it’s all the same, I have my client lib, my json objects and I’m ready to go, what can go wrong?!

My name is Yinon, API team leader @ Kenshoo and for the past 3 years I was focused on building integrations between our product and multiple publishers, In this session, I will cover the concerns, risks and how badly implementation of APIs might impact your product while providing specific guidelines for how to approach, manage and develop your product which relies on other [out of your control] services.

Rarely Normal or Normally Rare? Anomalies are by definition different and unconventional. A data point that doesn't follow the general rule. The difficulty with defining the nature of the unconventional resonates with all the steps of trying to find it. In this lecture I'll review the art and science of anomaly detection. We will explore classic and deep learning approaches to define and find anomalies with any domain as well as some motivations to look for them in the first place.

All of us have seen how image morphing looks like - how an image of a child is morphed into an image of a tiger.

Can you imagine what would word morphing be like?

In this talk I'll show you how you can use the combined power of AI and Machine Learning to implement word morphing yourself.

Has your organization recently stated how innovative it is? How many existing organizations still haven't gone that irresistible road? Many organizations, wether big corporations or small startup companies want to be as innovative as they possibly can. That's what's hot nowadays. However, the actions taken aren't always as innovative as they think, and they might actually cause serious problems in the long run.

Machine Learning has become an industry standard when it comes to personalisation - everyone is trying to utilise data to improve the user experience and increase relevance of the content. But, do we really accept the supremacy of AI over our intuition without any validation?

In this talk, I will guide you through real episodes of my front-end development work, as I go through case studies to demonstrate the importance of critical thinking, experimentation and correct analysis of results. This helps to challenge well-known assumptions and open up to new data-driven innovative ideas.

As a software developer I have experienced first hand the difficulties of growing startups. Today as the head of development in a scaling startup, I’m successfully implementing solutions to some common challenges. Here are just some of the topics we will discuss: How to keep the early core team engaged while scaling and incorporating new employees to the team? How to put some formal procedures while scaling? This session is relevant for developers currently working in startups since you could improve your work environment. For developers thinking about working in a startup since you could get a very realistic point of view on how it’s like working in a startup. And even relevant for CTOs and founders since it will be unique point of view on common pitfalls and how they can be avoided.

One of the challenging aspects of building modern games is making it multiplayer. Can we use HTTP there? We can, but we will not get a real-time game. Can we use WebSockets / TCP? Of course, we can and we should! However, developing API on top of WebSocket is another challenge that no one wants to solve, but everyone does. Existing libraries do not give us a required variety of communication models. Today we will show you the way to tackle this - using RSocket. RSocket is an application protocol that makes developing any networked application as easy as pie. To prove that, we are going to build a multiplayer Pac-Man, and during the process compare different ways of making a multiplayer game.

Part of the software developer job is to find new and better ways to solve problems. Writing code using his/her wits, intelligent and creativity.

However, sometimes being too clever can lead to hard to track bugs, maintainability issues and impossible to understand code. Is all cleverly written code good code, or is it a problem just waiting to happen?

In this session, I will show you real world examples of cleverly written code. And show you how we can use clean code principles, refactoring and design patterns, to transform that code from clever code to good code, code one that your peers - and future self - would thank you for writing.

We already know that client-side is no longer just html/css, and running on single threaded JavaScript is kind of sucks, so why not using the background threads web-workers? In my talk, we will build an example of making our site much faster when using web workers. We will talk about when to use them and when not, and how to check if we did a good job.

With Spark being widely used as a map/reduce framework for various data pipelines, it lacked capability of defining intuitive user-defined functions (UDFs). Contrary to Spark, Pandas is one of the most prevalent data-processing python packages and while it works well locally, it cannot scale to out-of-memory computations. In this talk I will briefly discuss the introduction of Pandas vectorized UDFs to Spark, why they matter, and how it enabled us to create a scalable and dynamic feature-engineering framework, crunching through TBs of data with ease.

The 2 major challenges for frontend apps these days are maintaining a sane architecture, and good performance. MobX introduces a very productive way to manage a reactive state with a single source of truth, using plain objects and decorators. This talk will introduce the idea and concept behind MobX, and how it can completely turn your web development process around.

Have you ever received an FE application as a gift from your manger? You were so happy 😊! But when you opened it you didn't like what you saw!! You didn't like the design, you didn't like the implementation, the build processes are too long and not clear, packages are not maintained, and a lot of dead code, not to mention lots of bugs and the poor UI/UX. Then you said to yourself “why should I do this to myself”? Why work with bad infrastructure? It's like putting a band aid on a wound and not really solving it. Then, you see the light! Why not build it from scratch? But how will I do it? I can’t just revise the code. I need a gradual evolution, with the two applications running together. Where to start? What will the design be like? Build process? Communicate & routing between them?

Have you ever attend a long discussion, where people kept talking and getting nowhere? Have you ever had a team that just could not figure out their progress? A retrospective meeting that was all over the place? Are you looking for any easy and quick way to do it? We often talk about the transparency that agile practices enables us while on our daily work we often forget visualization. We just want to have a short chat and get it over with. Visual facilitation and other visualization tools can boost your discussions and processes effectiveness just by displaying the data. In this talk I will share with you tips and visualization tools I often use on a daily basis that you can start using right away.

UI interactions are about to significantly change for the better, and this time it'll be driven by Frontend engineers. For several years, web interactions had a very little innovation, mostly due to designers not having code skills that provide exploration flexibility. On the other hand, Frontend scene is flourishing with stunning interaction projects that playfully engage users. The problem is that those projects are usually applied as backgrounds or out-of-context activities, without any direct impact on conversion. We'll discuss the mindset switch for bridging the gap between the interactive art and practical web design, with some open source examples we made.

The less code in our systems, the better we are. Less to maintain, less to read, less to onboard. And tests? They're code. A lot of code.

But wait, we need tests. Tests increase our confidence in what we've built.

In this live coding session, we’ll delete a lot of tests while keeping our confidence high.

The Samurai adopted Zen and practiced their daily work to the level of art. They were the best warriors thanks to their ability to be fully engaged and present in the moment.

Zen theory emphasises the importance of meditation and daily practice as away to control your mind and understand the simple truth of our existence. Zen is a way of life, work and art.

In the past year I started to learn and practice Zen. Step by step I’m merging Zen practices into my daily routine as a young father, husband and developer.

In this talk I will give you a short intro to Zen ideas and set of practices which can make us to be modern Samurai programmers and our code a piece of art.

we all want to make smart decisions (based on data) but often end up going with gut feelings. In this short talk, I will try to cover: why it's so important to really on data, how to get to a data-driven mindset, cover some basics on what the organization need to do so and some quick high level tips

This year was the first time I felt I completed the ho so well defined(?) data science "life cycle" - what a ride!
In this talk I'll describe the interesting question we tried to solve predicting whether a task we present to a crowed through our crowedsource app will be successfully completed on time. Given the fact crowed is an ever changing workforce and our tasks are flexible I'll try to use this challenge to take you through the bottlenecks, valleys and picks involved in the cycle, from a data scientist perspective. Hopefully, it won't be just a support group for frustrated DS but also help other individuals (product, exec, dev, PMs to name a few) to understand better the process of working with and aside to data science teams.

A review usually consists of the customer’s opinion on various aspects of the product, the service, or the website itself. Extracting these opinions can help e-commerce store owners understand the state of their business and improve it accordingly.

In this talk, we will give an intuitive explanation on how NLP models work, and describe how we built our Insights engine for extracting topics and opinions from reviews, along with their corresponding sentiment scores (i.e., good, bad or neutral).

When I started using git 7 years ago, I was baffled by its mysterious jargon, and intimidated by the occasional French-revolution threat You are in detached HEAD. WAT?

I do like my head attached, so I kept the secret industry best practice: stick to add, commit, pull and push, and if anything goes wrong - copy the repo to another dir, pray, and apply some online black magic. This sucks. I decided to learn git from first principles.

In this session I will explain the git fundamentals: file structure, SHAs and objects, commits, refs and branches, up to the git interactive rebase. After this session, you should be able to understand git man pages, error messages, and repo states - without losing your HEAD.

Deep Learning brings with it enormous amounts of data, complicated experiment results and intense compute requirements. Decades of experience in moving code to production yielded best practices in engineering that have not yet found their place in deep learning teams. Over the past few years, the term “DevOps” has become an industry standard but where are the DeepOps best practices for AI First companies? In this talk, we’ll cover why most AI companies fail when it comes to reproducible training, data management, and deploying to production.

The tension between developers and product managers in our industry is a well-known secret. In this talk, I will analyze why this tension exists and give some tips and ideas on how to overcome it & create a more synergetic win-win relationship between product managers & developers.

Clash Royale is arguably one of the most successful and profitable mobile games. I’ve been playing it daily for over 2 years, with several unsuccessful attempts at quitting. I’ve watched kids and adults rage-quit, swearing never to play again, and eventually return to play with even greater addiction. There’s a secret sauce that can, when used correctly, transform mediocre products to a huge success. In this talk I’m going to share some of the ingredient of this sauce: on boarding, notifications, social aspects, rewards and of course - FOMO.

"We have a problem in production! does it reproduce in staging?" is a sentence often uttered in many companies. Underlying this simple sentence lie implicit assumptions about environments which often turn out to be false in our modern services and applications. Clusters and micro-services complicate matters and make production environments hard to reproduce and model, but when A/B tests, feature toggles and ephemeral containers, serverless functions and multi-tenancy are introduced one must conclude we no longer have a "production environment", but rather "production environments", plural. In a sense, every transaction experiences a unique unreproducible universe. In this world, what is the value of Staging environments? how does such understanding change our development processes?

Git challenges shouldn't be a trigger for a panic attack. Git gained its popularity because of its excellent performance and powerful options that anyone can harness and turn to a master of their domain. Empowered by the terminal, Git can turn into a powerful tool and solve many problems.

Amazon introduced EC2 back in 2006. Since then, additional vendors introduced similar public cloud offerings. As a result, new applications and businesses were created based on the new features, cost and business models that these public clouds are offering. The question of private vs public deployments keeps being considered in many new as well as existing applications. The simple logic of "this deployment costs X in private data center and 3X in public cloud deployment" is missing some of the key consideration like: agile architecture changes, flexible consumption models and will likely lead to wrong conclusions. This talk will focus on a several considerations, based on real life experience, on when it makes sense to use the private data-center and when to use the public cloud.

With deep learning and computer vision, there are many pitfalls and hacks to work around and debug. On June 30th, 2018, Andrej Karpathy, Director of AI at Tesla, tweeted a short list of first things to check when your neural network isn’t working.

In this session, you will see what these mistakes look like in code and performance metrics. Using a computer vision dataset and a PyTorch code sample - we’ll walk through each of these pieces of advice, test it and explain it. Expect a technical deep dive and a review of best practices when debugging a PyTorch computer vision experiment.

Original tweet at: https://twitter.com/karpathy/status/1013244313327681536

While Deep Learning (DL) is proven effective black box, without the ability to peek into the inner workings of our model, to understand and guide the learning process, we quickly reach a glass ceiling.

We tackle this with an adversarial game: A DL model that learns DL! While it can be generalized, we'll focus on NLP: Given an RNN that learns to embed text, we introduce an "Editor" model - by observing the RNN’s hidden states (without access to the input text) it predicts how editing each word will affect the RNN’s performances - forcing it to analyze the RNN’s perception of language and guide it’s learning process.

We'll demonstrate the Editor as a highly effective analytical tool that reveals crucial patterns in our model's behavior, dataset and dramatically improve generalization.

Whether you are a data scientist or a machine learning developer, you have probably encountered the following scenario: you have a new bright and shiny machine-learning model that you validated with a test-set and it produces great results. Yay! After the successful deployment, users start complaining about the performance of this new model and you wonder - what went wrong? In this talk, we will map 3 major reasons to explain why machine learning models perform so much better on the test set compared to production and propose ways to avoid this situation.

Congratulations - you have just been promoted to a leadership role! Now what?? Once the dust settles, you realize your new job is much harder than anything you've done before. Suddenly things look pretty different compared to what you thought it would be like. Is this really what you signed up for?

This talk is an introduction for new managers on how to ease into their new role. How to start this new chapter of their career. Obviously, there is no silver bullet, finding your own way just takes its time. However, there are some principles and guidelines that can support you through it. This talk summarizes insights I collected during my own personal journey as a young manager, as well as from the young team leaders I had the opportunity to coach, mentor and manage on my day to day job.

This lecture focuses on the core of Web Components, it will answer the What, Why and How we use it. For years, libs & frameworks in JavaScript tried to build components for reuse, but the solutions suffered partially or all from: capsulation, inability to import templates, reuse, no standards. Web Components is the solution, It solves all these troubling issues above for agility, to make the apps more flexible, reusable and stable. It modifies the development approach to be more microservices oriented.

Resolving actor's identity is imperative in many online systems. Misunderstanding of your actors' identity means that you may be confusing bots with people, mistakenly resolving an individual actor as several different other actors, mislead by multiple identities which are actually the same actor and many more. These misunderstandings may well be wreaking havoc in your analytics be it by wrong visualization or by introducing noise to your statistical models. We will learn a strategy to try and mitigate this problem and how this strategy fits into a broader security system.

In years one and two you learn how to code. Then you spend the next 10 learning HOW TO CODE. The difference? Code written properly is easy to maintain, can be read like a book and you shouldn’t need to study every line to understand what it is trying to achieve. Anyone can understand what you wrote, why you wrote it and its purpose. Like you would never need to refer back to a previous chapter in a book to understand what is happening, you shouldn’t need to jump around the code to understand what a particular component, or statement does. Code shouldn’t just work. Code should communicate the concepts behind the software. During this talk I would like to share some of the techniques I have learnt in order to achieve that starting at requirements, to design and implementation to testing.

Running a Highly Available Large Scale Systems is a lot of work. For the past 2 years, we've been running 100% serverless on a full production environment, serving customers worldwide. No VMs, no containers, no Kubernetes. Just code. In this session I will present why we decided to go fully serverless, how it helped us move faster than our competitors, where did serverless computing worked best and where there's more work to be done. At the end of this session, you'll learn how to run your next side-project, production service, or even your complete startup, without losing a night's sleep.

Have you ever noticed that every Reversim episode comes out with a full transcript? Have you ever found it useful? or thought what's the point? For the last few months I'm helping in creating the transcripts for Reversim episodes, and found it to be a very interesting experience, with many benefits, some of them are far from obvious, which could be useful for any professional who's looking to expend her or his knowledge (aren't we all?). On this quick talk I'll review some of those aspects, and highlight some key takeaways any of us could use.

As a big data IoT platform we need to provide our non-technical customers with insights on their data.

Terabytes of data is streamed into a perfect physical model with continuous data pipelines aggregating it in the exact shape we want - and reads are fast!

Exposing it with - yet another API - to fetch some data brought us to an API hell!

We've looked for a concrete mechanism for analyzing, manipulating and querying data. That is, we needed a query language.

In this session I will explain why we chose building our own Custom Query Language and deep-dive on how to develop a blazing fast Query-Language interpreter with a live demo.

As a 16 years old developer (Gen Z), and before I'll become part of the official grown-up high tech community, I would like to share with you my learning journey and development experience. It started at the age of 6, when I first wanted to build a virtual world (now www.tornado.world). I will show you how your kids can learn to develop and how I think, they should start (forget about 'CodeMonkey', I'll talk about that also in my session).

How I overcame the language (English) barrier, being a 6 yo native Hebrew speaker.

Learning security through hacking Aroma, Cinema-City and some other SQL injections (including taking down my high school site after retrieving all students' info) I will also touch some legal aspect, technologies I used, revenues, development process etc...

Reactive Extension (Rx) libraries are everywhere, with the promise that they will make our code more declarative, less prone to bugs and overall simpler. But there is a price to pay for this simplicity – the Rx learning curve is notoriously steep. So before we jump head-first into yet another paradigm that promises to solve all our problems, let’s figure out what exactly there is to gain. In this talk we’ll look at some common front-end problems, and their proposed solutions in RxJS, and try to answer the question - does FRP really make our lives easier?

“Work Hard Play Hard” is a popular (and overused!) phrase among many hi-tech companies. But high salaries, game rooms, and Omer Adam are not all that matter in life -- or so I'm told... Becoming a dad changed my priorities, and made balancing my work life and personal life more important. In this talk I’ll share some of the things I do that have helped me find that balance, and try and answer the big question - am I still as valuable to my company and colleagues, despite spending less time working?

Battle tested tips & tricks for conducting postmortems/retrospectives. In the last 2 years we completely changed our learning culture at Trax Engineering. We would share with you the best practices we learned along the way. Practical ideas to switch from blame games to ace games.

A few months ago, I've worked on a POC, where I needed to find an appropriate API for implementing an infrastructure for Push notifications, which has direct relation to the Real-Time application arena. Due to a tight schedule (it always is), we needed to consider whether to use tentative solution using the well known Polling mechanisms...or to take the Red pill and - go deep to the rabbit hole. Wanna guess which one I took?

In this talk I'll cover the Real-Time apps evolution, the common underlying technologies to create Real-Time web applications today and then...giving a deeper glimpse into our Web-Sockets solution. Join me for helping you take the right choice yourselves.

We wanted to enable our end users to write and execute their own code in our platform. 
Here enters...
 Function as a Service (FaaS). It enables developers to instantly write, deploy and run their code with absolutely no DevOps concerns.

With one click you can instantiate a function that serves millions of requests, scaling up to thousands per usage.

Discover the HOWs and DOs… - Functions lifecycle. - How an FaaS instance works.

• Elasticity and scalability
• Cold-start and more…

Join us in a journey, raise the curtain and understand how a serverless FaaS platforms works using tools such as Kubernetes, Knative, Istio, GIT and more.

Deep learning technology and generative neural networks have just started the age of creative machines. We will review the latest state-of-the-art applications of generative deep learning algorithms in image processing, text generation, design, creativity and art. I will also exhibit my own digital artwork and perform live demos.

Medical Imaging AI uses deep learning to assist radiologists in recognizing findings. Zebra Medical Vision is aiming to develop algorithms that "find the Zebras" (why we are called Zebra?) Testing our algorithms performance is subject to many challenges, specially when tagging  significant amount of data requires enormous efforts.

I will present two use cases, in two algorithms where we developed a distributed dockerized reproducible pipeline system, enabling running multiple instances of algorithms parallelly.

We used large scale inference to evaluate emulated T-score, that measures lumbar spine density from CT scans, and Pneumothorax identification, a rare acute finding from Xray scans. Results of analyses done on these runs helped us evaluate performance during development process.

Internal company hackathons can be incredibly valuable. They're growing in popularity and for some great reasons.

If your organization hasn't run internal hackathons before, getting started can be a challenge. How do you explain to Product, or VP R&D, that you're about to spend at least a day working on anything but their features & bug fixes? What can you do to make sure that the teams start the day motivated and ready to work on worthwhile & interesting projects? How do you structure the day in a way that'll be fun & memorable, but still give people a chance to build something they're proud to demo?

We'll talk about what I've learned running hackathons for different organizations. I'll share my successes, my failures & everything you need to know to run your first company hackathon.

Deployment of large ensemble models can create large overhead on memory and disk space with energy consumption as a derivative. One solution is to reduce the quality of input data or the rate of inference. In this session, we will speak about a set of tools to make our trained ensemble networks smaller in size and in memory, enabling them to run on cheaper hardware or on edge devices.

Viz.ai uses AI to analyze CT scans of stroke patients to facilitate faster detection and treatment. This is a performance critical mission where any failed processing may result in harm to patients (1 minute delay equals the loss of about a week of healthy life to the patient). We leverage AWS GPUs to deploy our deep learning algorithms, and use dynamic allocation to control costs as the traffic changes throughout the day. Ensuring the health of a AWS GPU instance is critical.

In this postmortem we will tell the story of how we detected a faulty GPU hardware on an AWS instance, and how we now automatically detect and mitigate such GPU hardware failures to ensure 100% uptime.

What actually happens when you click your mouse in your browser window?

In this talk we'll take a deep dive into how our web browser works and what actually happens when a mouse is clicked. We'll take a deep dive into what browsers do, how events work internally and why.

We'll go through how web automation tools like Puppeteer and Selenium simulate user actions, when that goes wrong and why.

Ideally at the end of the talk developers would have a much deeper understanding of how browsers dispatch events and could utilize that knowledge to write better front-end code.

Being a part of a fast-growing company is an amazing experience but a challenging one as well. In a reality where your engineering doubles itself every few months, You experience an interesting side-effect that at any given time roughly 50% of your team is new.

So how can you build a codebase that can support a fast-growing company? I believe it’s all about empathy. It's about writing code that guides developers to make the right choices even when you are not by their side. It’s about making context visible in your code, it’s about helping developers focus on ‘how’ instead of ‘where’, ‘what’ or ‘why’.

In this talk, I’ll share some principles I’ve learned along the way on transforming our codebase for growth.

As developers, we sometime hear the advice to "go learn a new programming language every year". But why? In this talk, I will suggest that other than keeping up with current trends, exploring new languages makes us better coders in the ones we use today. You'll hear about our own experience at Riskified, where we found that working on greenfield Scala apps helped in levelling up our existing Ruby codebase. Come and get inspired about geeking-up and exploring new territories! Also, you might figure new ways to convince your boss that reading blogs and books at work is, in fact, time well spent!

Everyone does IO Everyone cares about performance. Linux exposes a lot of information about the inner working of its block devices.

But until now, the only way to access this information was to run the blktrace command line utility.

In lightbits.io, we wrote a python library that enables programmatic access to this information.

At ebay we operate at scale. 3-4PB of application logs per day, ~300K compute resources (BM, VM, containers), 1.5M system metrics per hour. This requires some consideration and planning into how to find the proverbial needle in the haystack.

This session describes some of the challenges we faced, how we tackled them and how we gained real value from application monitoring (in production and all other environments we have).

One day I desperately needed to pass a mandatory web security e-learning. Every stage ended up asking for a flag that could be obtained by completing some lab exercises. The problem was ... somebody left the office or company, turned down some boxes and the lab servers were not available... Problem? Rather a challenge! First thought - live off the land! This live demo shows how to take advantage of Developer Tools embedded in web browsers including Chrome, Firefox or Edge to perform rapid static and dynamic analyses of JavaScript. The example will focus on passing the annoying e-learning, but there will be also a quick story shared, how the same method allowed me to score some best-dad points helping my son to impress his girlfriend

Creating a real-time image processing filters is difficult, let alone on a mobile device. In contrast to many other real-time data manipulation tasks, the amount of elements that need to be processed on every frame is huge, and the operation can rarely be deferred to stronger, offline systems. At Lightricks, we've spent the last 6 years optimizing our iOS and Android image processing engines to provide a seamless user experience.

In this talk I will present a case study writing a gaussian blur filter. We will begin from writing a naive CPU implementation that performs poorly on large size images and gradually work our way to an efficient, scalable GPU implementation that runs in real-time.

No prior knowledge in image processing, GPU wizardry or performance optimizations is required!

SRE - site reliability engineers is the term for proffesionals who deal primarily with handling production incidents. in this talk we will several "light bulb" moments that happened to me, when working with SREs, in which I realized how point of view of an SRE is different than that of a coder.

Taboola’s R&D had expanded from ~50 developers to 250 in the past two years. With a fast and effective CICD pipeline we are rolling out releases on a daily basis. Because multiple developers commit new features to the same code base, we had to have a smart error detection tool to prevent a bad merge from happening. This led us to develop an in house Error Detector Mechanism, which is sensitive to the nature of our data.

A development error can be DC specific, and peaks of errors can appear in certain rush hours of the day, certain days of the week, etc. Taking into account these factors, our constantly updating class-specific baselines can detect errors fast, with zero effect on performance, saving money and post mortem analysis effort.

We invite you to see how it is done at Taboola.

Early stage startups may find it hard to to hire developers. They lack the money for a decent salary, they sometimes need to build the very basic foundations of their code base, and the technical co-founder may end up doing tons of other stuff that are not development-specific tasks. My short talk will share my experience as a co-founder struggling to build a product with very limited resources. I will discuss more creative solutions like hiring interns (where to find them, how, what to expect) and using remote developers as contractors

We will be examining an extension I’ve recently built for VSCode called Glean. Glean, is an open-sourced refactor toolkit for React apps, and has gained popularity lately, with over 24K downloads. https://marketplace.visualstudio.com/items?itemName=wix.glean Blog: https://www.wix.engineering/blog/finally-a-react-refactoring-tool-introducing-glean

I'll argue that you should stop commenting your code and will show how it'll make it better. We'll see examples in code with and without comments and notice the improvement. We'll also briefly talk about the special cases when it's OK to write comments in your code. Code examples will be in a programming language but apply to others as well.

That’s it. You are a system, lead or chief architect. You really want to make a difference in the organization. But how? Designing more and more features, building architectures for more and more infrastructures, is not enough. After doing it again and again, I understood that lead architect goal is mainly focused on the architectural vision. But I like to be practical, and setting a vision is not enough. I want something concrete, tangible and measurable. In this session I will present you the “well architected” pattern. This pattern includes 3 sets of standards and guidelines: Software architecture; People & Processes; Quality. We will drill down to the pattern, and I'll show you how to use this tool in order to set an "architectural vision" that is practical.

Deep learning applications are used across all life domains, and inference applications are getting deployed everywhere from data center to edge devices. This brings an appealing opportunity to create an accelerator to optimize inference flow. In this talk I will describe the software challenges we faced in building such an accelerator

Continuous Integration has almost become a de facto standard in most organizations nowadays, however we are all still not giving CI the proper emphasis on our day to day work. Why invest in your build times when there are sexy things like Continuous Deployment and Kubernetes? Now that we have a plethora of CI tools and SAAS offering, are there still things we can do differently?

In this talk I'll delve into some of the history of CI that's still relevant, go over very common mistakes and how you can level up your CI game with very little efforts.

We are engineers.

Or at least rationale, realistic and methodological creatures. We are great at optimizing CPU cycles, load times and conversion rates -- but what about our own career path?

During our professional career, we are given countless opportunities to steer it towards its optimal course.

But what are we optimizing exactly? What are the choices we are given? What is the optimization function? What hidden factors are part of it? Are there known "right" or "wrong" choices?

Let's optimize!

Keeping your code organized and versioned is of obvious importance to software developers. It saves us precious time. The machine learning (ML) field is only beginning to catch up. This stems from fundamental differences in workflows, requiring different tools.

In this talk, we will review these counter-intuitive differences and the state of the ML world today. We’ll introduce DVC, an open source command line tool that complements Git and enables organizing and versioning of machine learning projects, in order to fill the gaps required by the ML workflow. We will show a case study, and attempt to encourage the growth of an open source ML community in Israel.

As we build products and features, one of our main goals is for our users to actually use them. One of the best ways to increase product adoption is to create Passive Impressive features - those that remove the need for the user to opt in, but without being intrusive or “Passive Aggressive”. Or in other words - opt out features that encourage users to ask “Why Shouldn’t I use this?” instead of “Why Should I?” In this session, I will talk about why we should focus on “opting out”, why it doesn’t apply to everything, and give tips on how to increase the focus on this in your organization.

Taboola has >250 developers creating over 20 releases a day. We have almost 700 frontend servers in 7 data centers serving over 500k HTTP request per second. Going from QA to production means exposing a new feature to 1.4B monthly unique users and up to 500K HTTP requests/sec which can be frightening.

In my talk, I will demonstrate the concepts we use in Taboola to tackle the above challenge. How and what technologies we use for our metrics, alerts and Canary deployment. How we encourage developers to take end to end responsibility and know the business.

I believe that working the way we do make me a better developer. It is helping me be more creative and responsible, take bigger risks and most importantly enjoy. I will be happy to convince you to try and work as we do.

Deep Learning models have been gaining increasing attention in the recommendation systems community, replacing some of the traditional methods. The sparse nature of the problems and the different inputs types offer unique challenges for feature engineering and architecture planning, in order to balance between memorization and generalization.

During the past 2 years the algorithms team in Taboola moved all of our algorithms to DL. In this talk we'll share the lessons we learned doing so. We'll talk about building NN with multiple input types (click history, text and pictures); feature engineering in DL; capturing interactions between features; and the way modelling decisions are related to system engineering and research culture.

Containerised deployments have rapidly become an industry wide standard.

We use container engines such as Docker to develop and deploy our applications on a daily basis. Many businesses depend on this technology to survive - but do we understand how they work?

This talk aims at demystifying concepts and terms such as cgroups, namespaces and “lightweight virtual machines”.

By writing a container from the ground up we will discover that a container is simply a sandboxed Linux process.

Audience members will walk away with

• Understanding the underlying concepts of containers.
• Learn how to implement a container from scratch.
• Understand the pros and cons of this technology and what it can do for your startup / business.

Distributed systems are hard and unfortunately often less perfect than we would hope. This talk shows a concrete example: Elasticsearch's cluster coordination system, called Zen Discovery, got a rewrite in version 7.0. Starting from a formal model, the coordination layer was rebuilt to address multiple issues discovered over the years. We discuss the main improvements of the new implementation: Master elections are much faster, the infamous minimum_master_nodes setting has been removed, growing and shrinking clusters becomes safer and easier, and leaves less room to misconfigure the system. Let us join the new, more Zen way of cluster coordination.

Web development changed dramatically during the last years. With the enormous amount of JavaScript libraries and the HTML5 standard, today it is much easier to create web apps. But when you want to develop your web app what framework should you pick? React? Angular? Vue? In this session we will ask ourselves if we really need a components framework.

One of the critical factors for development velocity is software correctness. Our ability to develop and ship new features fast is bounded by our ability to validate several aspects of the change:

• Does the feature meet the requirements?
• How does the feature affect existing code, and how can it affect the production environment?

In this talk, I’ll focus on testing environments: why developers need a self-serve platform to create a full functioning environment on-demand, how such environments should be managed, and how can one restore part of the lost velocity. I’ll cover an internal system we use at AppsFlyer called ‘Namespaces’ that addresses the issue with the help of Mesos.

When my manager offered to become an architect my big concern was..... to be an architect mean that i will stop write a code as a developer. I'm so happy that my concern didn't turn to be a true.

Having say that.... Many think I'm no longer fit to be a developer In this talk i will try show that you can be an architect (or a manager) and still be an hands-on developer

We will do it base on code kata (like a 99-bottles Kata) I hope that after the talk the opinions a about architect as a developer will be with me

Choosing how to handle your data is not an easy task. Every database and data processing framework will try to tell you it is the best in the market and can solve all of your problems. But when real life strikes, you learn that every solution has its pros and cons and you might find yourself struggling to fit your data into the right model. In this session, we will tell the story of how the data architecture of Otonomo evolved during the first three years of the company. You'll understand why we made some technological and architectural decisions based on the ever-changing nature of a growing startup. Having all the tools in the belt is not enough. You have you know when to use them.

הרצאה בתחום טכנולוגיות שיווק לצוותי פיתוח ומתכנתים. הסקירה תציג למתכנתים את המהפכה הטכנולוגית בעולם השיווק הדיגיטלי ותיתן להם כלים פרקטיים למלא את התפקיד המכריע שיש להם בפעילות השיווק הדיגיטלי של החברה. החל מהתפקיד הקריטי שיש למתכנתים על קידום האתר במנועי חיפוש (SEO) וקמפיינים ממומנים וכלה בהשפעה שיש למתכנתים על נושא המדידה והאנליטיקס של האתר. למה לצפות?

• סקירה של SEO טכני והגורמים הטכנולוגיים שמשפיעים על קידום האתר בגוגל.
• כלים לניטור ואופטימיזציה של הסריקה של גוגל.
• סקירה רחבה וכלים פרקטיים בנושא מהירות האתר וההשפעה על ה- SEO (כולל סקירה על AMP)
• סקירה פרקטית בנושא אנליטיקס ובחינת הגורמים שיכולים להשפיע על נתוני האנליטיקס של האתר.
• נתונים מעודכנים לגבי ההשפעה של Java Script ו- AJAX על האינדוקס בגוגל
• הסבר על נושא ריטרגטינג וטירגוט קהלים וההשפעה שיש להיבט הטכני על הצלחת הקמפיינים.

We all work with open source projects and wish to report bugs or missing features that we need to be fixed. Maintainers of open source project frequently context switch between reports from completely different production setups and workloads, causing a long wait until those are handled. In this short talk, I will illustrate the right way to open a bug report for an open source project that will help the maintainer, rather than alienate them, and speed up the resolution!

Even if you are familiar with React Native, do you really know how it works? Did you know about the bridge which is the secret sauce of React Native and makes it work on every platform and gives you the option to create a Native-Like apps? In my talk,I am going to reveal that there is no magic in React Native. There is a real smart mechanism that lets the JS code communicate with the Native code. A major part of this mechanism is The Bridge which is written in C++ (wait..what??) and mapping between all you app modules and even lets you create custom ones of your own! In Addition I will talk about its new architecture. If you are a curious React developer who believes that knowing the internals of a library makes you a better developer, join me to the journey Under The Bridge.

Writing secure code is hard, and that’s why I love the idea of security tests. I’ve spent countless hours researching tools, integrating them into the pipeline, and help developers use them. It’s all based on one big assumption: security tests can detect real, juicy vulnerabilities. But have we ever stopped and questioned this assumption? I didn’t – until I started preparing a talk about how security tests can help people write more secure code. The point was to see how many vulnerabilities we can find by scanning a vulnerable application with security testing. But you’d be surprised to find out how many were actually detected. In the session, we’ll find out how many security issues we can detect in OWASP Juice Shop by using various security tests. Ready to put security tests to the test?

Community detection algorithms try to partition a graph into communities with high connectivity inside each community and low connectivity between communities. In this session, we will start with an overview on popular community detection algorithms, optimization goals and parallelization methods. Then, I’ll present the community detection pipeline we developed and applied over probability-weighted identity graph in Oracle Data Cloud.

The web is a crazy place! In this talk I will go over some weird and dark aspects of the JS eco-system, highlighting common security issues and pitfalls which we, as web developers, can learn to protect against. I will start by going over some wonky NPM packages that made it into the JS ecosystem, including: “NoCode”, “IsThirteen” and “JsF**k”. Then, we will revisit a couple of stories that are now part of web history, including #SmooshGate and recent NPM security issues from 2018. Next, I will go over some fun pranks using CSS to mess with people. Lastly, I will present some techniques one could use to make the most annoying website possible, stressing some common security pitfalls which front-end developers can learn to identify and block, such as the infamous “window.opener” property.

👾 Meet WebVR and A-Frame & Let's build a VR game together - Adi Polak and Uri Shaked 👾 WebVR is an emerging standard that enables web developers build Virtual Reality experiences using the familiar technologies we all love - HTML and JavaScript. We will show you how you can create a fun VR game, from scratch, that will run on multiple platforms - Oculus Rift, Google Daydream, Windows Mixed Reality and even on standard phones.

We will live-code everything (with some help from the audience), and then call some brave volunteers to play the game we just created and compete against each other. Who wins?

2147483647 = Integer.MAX_VALUE We have almost reached this number of entities in our DB. we found it a few days before overflowing. It could have been Outbrain Y2K bug, but it wasn't.

In this talk, I will describe the problem that we had, the action we took, the results, and the conclusions

We all have them - admin pages or back-office tools - but nobody really wants to invest in them. This is a story about how two human errors in a single admin page - independently and within 24 hours - almost lost us a huge customer. We’ll explore together how reliable admin pages may turn dangerous as organizations grow, and how a change in our mindset allowed for an improved product with happier users. This talk attempts to sharpen your instincts - you’ll understand when an admin page is actually an application in disguise.

In this talk I'll deep dive into building a machine learning framework over Spark for content recommendation, and show how we use Elsaticsearch-based serving architecture to recommend content, at a scale of billions of recommendation requests a day, and under strict latency requirements of several tens of milliseconds per request.

By now, you're probably thinking: "This is yet another CI/CD session". Well, its not - at least not the ones you are used to.

Think back to the first time you heard about Continuous Deployment. You immediately thought that's a fantastic idea and a right fit for your company. That as a long time ago. So why aren't you doing Continuous Deployment yet? In this talk we'll cover the technical, but most importantly the organizational and physiological barriers in adopting Continuous Deployment, and how to slay each one. After attending this talk you will have the tools to convince the stakeholders, detect bottlenecks and lead the way to a smooth, painless, automatic deployments in your company.

Automatic analysis of medical imaging is transforming medicine as we speak. However, it is no easy task - the medical imaging domain poses unique challenges in one's way. For example - privacy concerns make obtaining data and labels a major challenge, medical imaging data is not your standard RGB input, can be 3D or even 4D and can often be of huge size. In this talk I will describe some of the challenges we encountered at Viz.ai, how we tackled them, and how this work translates into saving lives.

One of about every 60 children in Israel has some form of autism. Some of these children require special attention and often need help navigating their school’s interior space to move from one room to another. After the “Yahdav” school had won the innovation competition with their brilliant idea, AT&T Israel accepted the challenge to implement an open-source solution for indoor navigation for kids. Numerous technologies such as GraphQl, nest.js and angular, as well as Unity and Kotlin, were used to develop this project with continuous integration while maintaining high code quality and speed of development. Collaboration between a large number of people made this project possible. This a story of success, technology and collaboration.

When we began working on a new system that will replace other, ageing, systems, we decided to go for the Micro-Services path. Our starting point was tough - There was no Micro-Services-Oriented knowledge in the organization, we couldn't use cloud-based product, and the schedule was quite strict. Now, a year and a half after making that decision, we can look back and analyze that decision. In this talk we'll discuss, among the rest:

• Why did we decide to go with Micro Services at the first place?
• What did the design phase look like?
• What kind of training did we have to do?
• What technology and platforms we decided to use, and why?
• And, of course, did it work?

Don't you just love creating a good, elegant, juicy abstraction? Extracting the ESSENCE of a problem. Coming up with generalizations. Debating their relative merits on social media. It's all so much more fun than solving these pesky bugs that keep coming back from QA. We are developers, dammit! Abstractions are what we do!

On our day-to-day, we often don't think about the level of abstraction we employ to solve a problem, or how the abstractions we choose affect the kind of solutions we reach. But what if we could think abstractly about abstract thinking? Let's explore the ways the mental model of the ladder of abstraction can help us choose our career path, solve problems, deliver clearer messages and help others grow.

very meta. so philosophy. much recursion. wow.

When talking about running scripts in production a lot of developers would say it is not recommended or even tabu. Looking at the other side, the users that use software every day in production like IT, they run scripts all the time. DevOps is mainly based on running scripts, everywhere all the time So why as developers, we don't recommend it? What are the advantages of (if any) are there for running scripts? Why should we care? What can we do to help?

Bugs vs Features is a daily rat race and a reality for devs everywhere. But haven’t we had enough? How can we find the details that if noticed, would save tons of time and make dev teams happier? That’s what this talk is about. I will cover:

‘Rebuild’ Button An “innocent” button gone rogue. Why do we keep pushing it? And should you care?

Jenkins/CircleCI/Docker/AWS sucks Let’s move past criticizing our tools and juice them for better performance.

Debug Hell The value of monitoring tools, and why each minute of not using them will turn into hours of debugging.

Numbers Don’t Lie How much of our time simply goes down the drain? How can we get the metrics to find issues? How do we know we’ve fixed them, and how to convince the MGMT these points are crucial.

Scaling is one of the biggest challenges in today’s startup world. We can never think ahead about all the features, the scale and the challenges that the success of our products, and the growing user base can bring along with it. In this talk, I’ll show case some of the challenges I had to overcome, and suggest a new mindset of building scale-oriented products. Why is it so hard to scale from 5 users to 100 users, or to scale from SMBs (small/medium businesses) to enterprise customers? In this talk, I’ll try to convey the mindset and tools to help with this massive change.

What if I told you that not all managers know everything?

This lightning talk shows my feeble attempts as a technical engineer manager to not be overwhelmed by my smart minions. In addition, the session talks about of how a manager can bring value even without taking a pause to get a P.hd in every field the team members are experts in.

API gateways are a common practice - usually the "public face" of your internal system & are served via one or more backend services. Besides providing a uniform API, they also facilitate a standard way of authentication, permissions, versioning & much more. What if we could gain some of those benefits when we build our web applications? What if we could compose our app from multiple agnostic parts, each with its different underlying technology & version, thus, enforcing a global authentication flow without rebuilding the whole system?

This talk will show you how we took the core concepts of an API gateway & applied them as the base architecture for our web apps, & scaled to 30+ apps in production while sharing libraries of various versions, managing a global state, routing & more.

Something changes in companies when they grow above 150 employees. But it is not only possible to achieve exponential business Growth with a steady head count, it’s also more fun, engaging and effective. Instead of hiring, focus your efforts in Growth dev, marketing and the happiness of your employees.

Imagine that one day, your beloved and supporting backend partner decides that you should start seeing other API providers. “It’s not you, it's me...” That’s what happened to us, our backend was going to get deprecated and a new backend was about to be written in a way that supports multiple API clients and not just us. Our endpoints will no longer be supported, our domain entities will no longer exist. How could we go on?Do we need to rewrite our entire code base to use this new backend? Is there anything we can do to ease the pain? Luckily, there are other fish in the sea, and in this talk we’ll discuss the node server we created that helped us bridge between our existing client code and this new multi-clients backend.

What do you think about when you hear “Tech Debt”? Most people think about bad things which must be avoided at all cost. Bad design, no tests and unmaintainable code are just a few examples. When I hear “Tech Debt”, I think about managing risk and opportunity. I think about a powerful tool which, when used correctly, allows delivering great products. I believe learning to use tech debt in a smart way (vs. “cutting corners”) is difficult, but is a key factor in becoming a mature engineering team that ships successful products. In my talk, I will explore common misconceptions about tech debt, what types of it exists, and how to leverage it to achieve your product development goals. I will also cover the fact that tech debt must be payed back, and share practical tips on how to do it.

I would like to talk about Sound Design in applications & games and to share from my experience the following topics:

• Tips on how to easily create/edit better sound effects and music for your apps.
• Tips on how to work and talk with sound designers.
• What is Dynamic Sound & Adaptive Music, what does it have to do with your app, and some amazing tools to use in order to implement that in your app.

1.5 year ago I decided I want to choose when I want to work and where I want to work from. So I found a remote job and bought a one way ticket. At this talk I'm going to share my experience of how I did it, what are the benefits and the difficulties in working remotely, how to work with a remote team and is that the right thing for you.

Everything that has a beginning has an end. EVERYTHING including the programs we write and run. How can we prepare to this final goodbye? Turns out that In order to die well, our programs need to live well - or at least have a decent architecture. Also don't try to cut a deal with death, zombies are the worst

While scheduling a limited number of data workflows is a generally manageable task, scaling to hundreds of data workflows with dependencies and diverse job types, requires a substantial customized engineering, complexity, and overall expensive resources. Serverless-based architectures offer an alternative to traditional resource management.

Tomer Levi explains how the data engineering team at Fundbox uses AWS StepFunctions, Docker containers, and Spark to build a live serverless data orchestration platform. Tomer will further describe AWS StepFunctions state machines, their limitations, and how to overcome them by building a custom job scheduling and dependency features. Finally, the talk will illustrate how resource bottlenecks were overcome using Docker containers and AWS Fargate.

As we mature as developers, we develop our personal style, best practices, dos and don'ts in regard to our craft. One major part of that craft, in my humble opinion, is testing.  In this talk, I'd like to present some of the guidelines I personally use to write better, readable and maintainable tests.

So, everyone is doing GraphQL and you want to do it too!

You created a POC for a GraphQL server and it looks great! It works, its cool and your client devs are happy. You want to graduate to a production solution but... then you start running into real life problems like authorization, performance, boilerplate code etc.

We were exactly at this point a year ago, and in the process of solving these issues we learned a lot. In this talk I will share with you what we learned and use this as an opportunity to dig deeper into how you can use GraphQL's strong points to solve common problems.

So, if you are curious about GraphQL and want to get a deep dive into this world or if you are a seasoned GraphQL veteran and want to get another perspective, this talk is for you.

Interactive programming using notebooks (such as Jupyter), is probably one of the most productive ways to write software. More specifically, it’s an amazing tool to use when you’re learning new things (like with data research), and specifically when learning a new shiny technology. I think this is kind of paradoxical because in my own experience most libraries do a poor job of making it easy to use them in an interactive environment. This is a story about my own python library that did a very bad job as an interactive tool, and how I solved it by creating custom auto-completion tools. We’ll go into why and how you can use the same methodology to improve your own favorite libraries, and even dive into using the same ideas for non-interactive development using VSCode language servers.

Data science and machine learning have become very popular recently and some of the “magical black-box” methods such as deep-learning became common buzz-words in tech industry slang. However, while moving from data science towards leading people, you suddenly realize that there are things that you cannot apply directly from online courses. To make this transition smoother, I propose to “transfer learn” machine learning jargon to management. In this talk I’ll share a few machine learning concepts that can help you understand team behavior better and explain how to deal with the “dropout” of your team members and avoid “overfitting” your approach to team leadership.

As software architects, we usually have a few parameters when designing a solution for an upcoming feature. Those parameters might be delivery time, maintenance and extensibility. But we often forget about one crucial parameter: cost.

When your team keeps cost in mind, you start exploring new tools that you didn’t even know existed. So in this process, you learn about new technologies and you start thinking more about efficiency rather than sticking with what you know.

In this session, I'll explain how we adopted a culture of considering costs, how we track the cost of our production environment and how in three days we started saving $25,000 per month. I’ll also discuss how thinking about money expanded our team’s knowledge in new and up-and-coming technologies.

Spun off of the eXtreme Programming movement, Test Driven Development has taken the software industry by storm. From a highly controversial approach, TDD has become a ubiquitous industry practice and the poster child of software engineering maturity. But is it all it's cracked up to be?

Severely misunderstood, argued about vehemently and lacking in critical research, the popularity of TDD is such that it's barely challenged anymore.

In this presentation we'll take a journey through the hype and into to the essence of TDD, observing its shortcomings, hidden costs and unfortunate dogma.

In the last decade, more products and teams choose a front-end application as their UI layer. Under the umbrella of a singular front-end system, comes a wider dependency on interrelated packages. The packages management is usually complex and time consuming. However, we usually address it as unavoidable. My talk main notion is to introduce the Package Coupling and Package Cohesion principles, as a very rewording architecture patterns set to cope with those challenges in front-end systems. I will suggest a set of guidelines based on my interpretation to those principals, for making a system stable, maintainable and testable, and allow decoupled deliveries, without hurting reusability. I will enrich the discussion by supplying a practical examples of CyberArk architecture evolution.

Haggling is a negotiation game in which pairs of players try to reach an agreement on how to split a set of assets, while striving to maximize their score.

In Outbrain we used haggling as a coding challenge in the form of a game / hackathon which swept up the entire R&D organization; keeping everyone playing, strategizing, coding, hacking & learning - all in two weeks of pure energy and fun. The winning team was interestingly comprised only of woman engineers. Our secret sauce? Come and hear!

In this talk we'll tell you more about the haggling game, how we leveraged it as an R&D organization wide learning activity, and how we won!

The field of Data Science is currently on fire. Many people want to work on something related to it. But what does it mean to be working in this field? I am going to show you a part of my daily work as a Machine Learning researcher. Specifically, how do you approach a research paper and implement the described algorithm? In this talk, I will share practical tips and advice I collected during my years in academia and the industry. I will present examples how they can be applied on deep learning algorithms.

Functional programming has many advantages in writing clean, maintainable and scalable code but one major problem - people have a hard time understanding it.

I'll use my imperative background from writing Java for many years with my functional experience from Scala and Javascript to come up with motivation and implementation principles that anyone can understand and use. even Java developers (I use them myself).

I'll even manage to do it without any mathematical formulas or examples from Haskell

The lecture will be valuable for people that have zero experience with functional programming but also for people who have some sense and wants to understand it better

In the past 10 years as a front end developer, I found myself creating internal management tools for RESTful APIs over and over again, while in fact, the usage of these tools remains the same: creating, updating, and viewing data based on an API.

I had enough of it, so I created RESTool - a generic UI tool for managing RESTful APIs.

In this talk I'll present how to configure, run, and use RESTool, and what caused it to jump from 0 to 170 stars on Github in couple of months.

Link: https://github.com/dsternlicht/RESTool

Randomization is an important tool in any developer's toolkit, but do you know how it actually works? Did you know some approaches can be more "random" than others and what those implications are? In this talk we'll take a deep-dive into pseudo-random number generation in JavaScript — how they are generated, how to make more cryptographically-sound random numbers with the Web Cryptography API and why we should care. We'll also explore practical and creative applications of randomness in building software.

When AI/ML model goes to production - it immediately starts to have its own life and meet the reality of the production-mess data streams. Without the proper monitoring - a big mess is just a matter of time, letting the algorithm take decisions on GIGO basis (Garbage-In-Garbage-Out). Looking at the traditional software/infrastructure KPIs will just dont do it. In this session, we'll share a few of the most important KPIs to monitor when you launch your AI/ML in a production environment.

Being a good developer comes with a never ending pursue of the latest technology. Learning never stops making it difficult to find the desirable work-life balance we are all seeking for.

Imagine a concept which assists you to be always updated with the latest trends and technology including hands-on experience, yet also makes time to actually have a life !

If you’re only thinking technology, think again! Social interactions are equally important to create a more stimulating, productive and friendlier workplace environment.

In this presentation I will show you the environment I work in which implements exactly that. We will cover the way to set your work environment to be best technologically and socially.

How do you know your feature is working perfectly in production? And if something breaks in production, how will you know? Will you wait for a user to report it to you? What do you do when your staging test results do not reflect current production behavior? In order to test proactively as opposed to reactively, test in production! You will have increased accuracy of test results, your tests will run faster due to elimination of mock and bad data, and you will have higher confidence before releases. You can accomplish this through feature flagging, continuous delivery, and data cleanup. Only when your end-to-end tests pass in production will you know that your features are truly working.

All our software is a bundle of decisions. But despite the fact that some decisions can have a long-lasting, crippling effect on our app, we make most of them based on guesses.

But what if the best decision is indecision? What if You could postpone most of your decision until you gather enough facts in order to make the optimal one?

In this talk, we will learn why deferring decisions can result in a much better, simpler code base. We’ll see examples of decisions we think we have to take on day 1 of the project, yet, can (and should) be postponed and see how practices like Spikes and the right design of your application will allow that to happen

Building modern web apps require a lot of Javascript, which is necessary for making them intractable. Delivering JavaScript efficiently by splitting your bundles smartly, can shorten your time to intractability and overall load times significantly. In the session I will share how Javascript bundling work at Facebook, and how you can implement similar ideas in your web app.

Apache Kafka is a battle tested stream-processing platform. Combined with Kafka Streams API, it’s possible to build scalable, fault-tolerant and distributed event driven systems.

In this talk, I will describe the key components of an event sourced system, and how simple building blocks can be combined using the powerful and declarative streams API to deliver business value. I will demonstrate possible solutions for the challenges of eventual-consistency and the object-relational impedance mismatch.

As software developers, we are required to sit long hours in front of a computer trying to solve difficult problems. This way of life requires us to go meetings, reading emails, writing emails, writing designs, writing code and read Slack messages every day and list is never-ending. So, how can we Concentrate more on the actual work with all this disturbance? There are a lot of different ways to achieve concentrations and efficiency at work. From modern day prescription drugs to meditation I will try to show you all the ways to be more concentrated and efficient.

You know the one about the recruitment ad seeking an under 30 developer with over 20 years of experience in a specific FE language? What if I tell you that a FE developer with just 3 years of "relevant" experience, but with 5 years of experience as an architect - could actually do a better job, and improve the entire company's performance?

Multidisciplinaries not only have in-depth knowledge of various fields often making them the go-tos for integrations, they also bring methodologies that create groundbreaking innovations and creative solutions.

Come learn how to make the best use of your entire team's diverse experience, and bring your own irrelevant experience to work, in a way that makes you stand out in the crowd.

A common misconception is that data science is only about choosing the right model for the problem at hand, when, in fact, considerable time and effort are put into more fundamental challenges. One major bottleneck in machine learning is getting reliable labeled data to train the model on.

How to learn on partially labeled data? How to deal with a data set that may potentially contain miss-labeled observations? How to create a feedback loop on your model predictions?

These are some of the labeling challenges that we are dealing with in PerimeterX. Many of them don't have a single, clear solution, and can be approached from different angles. Our talk will review some academic approaches that discuss these issues alongside case studies from PerimeterX.

It's everywhere! companies are offering "data driven" culture instead of young and vibrant atmosphere. Statisticians are data scientists, ETL developers - data engineers, data architects appear everywhere (they were here all the time, just feel safe to come out to the light now), and executives are talking about "data assets". One graduate even described himself as "passionately data driven" in his CV!

Has everyone lost their minds?! (Sorry, "got their feature set mixed up")

In this talk I'll share insights I collected during my journey from "plain" software engineering to managing Kenshoo's Data Platform Group. I'll try to demystify the term "data driven" and what it actually means for a company. And most importantly, what to do when your CEO ask you to leverage the power of your data.

Looking at the ongoing risks on any company today, no matter how much time and money you'll invest in cyber security as CISO or any security practitioner, You always arrive to same conclusion, it's not enough to achieve high standard of security, you have to do it 24/7. In the talk we'll give a brief overview on different aspects of Cyber Security in the modern world. The "How": We'll share how to build an infrastructure for security researchers that will allow them to concentrate on business logic and writing hacker “tasks”. Using Docker and Kubernetes on Google Cloud. Lessons learned promised and a glimpse of the Hackers view, because it's always interesting to see how you look from the outside.

Only half of a mobile app's users use it more than once. Maybe they just got what they needed, but often they leave because they’re overwhelmed, not sure what to do or think they can’t get what they came for.

Good onboarding and introduction can drastically improve this, but it’s not easy to make it work for everyone. Some users like to be guided and hand-held, while others just want to explore. How can we support both types of beginners? Can we avoid a Clippy situation, an onboarding process that detracts more than it helps?

I spent 2 years at the MIT Media Lab investigating how to make it easier for kids to get started with Scratch, a coding environment for kids. In this talk I’ll share some key points learned along the way and demonstrate how to apply them in any onboarding process.

One of the challenges that development teams face across the world, is optimising a product to achieve fast load times while maintaining performance and quality. Monitoring performance is a key aspect of an engineering team, though in some cases, using third party services, or imprecise indicators (like network issues), might affect performance.

Introducing a runtime, deep analysis tool named “Load Time Screen” which visualises the loading phases and adds transparency to the loading sequence, lets us investigate reasons for slow load time of a specific instance. This allows support teams and R&D teams to get better information from the client and identify issues.

In this talk, I will demo and elaborate on this technique, why it’s important and how to implement it in your product.

Speed vs Stability that's the basis of an age old discussion. Recently with the release of the research in the Accelerate book we have actual scientific proof that it's not this or that, rather it's both. In this talk I will go through the main points researchers present and explain what makes an elite organization 2555x times faster and 2.5x more likely to succeeded than a low preforming org. Covering topics from deployments, org structure, feature flags, monitoring and more we'll review how to shift your org to be the BEST.

In this session, I will describe the fascinating world of voice biometrics and how we use it at NICE to address many use cases in our portfolio. I will explain how it is generally performed, main algorithmic approaches, the main challenges around such technology, and evaluation techniques.

Building distributed systems is notoriously hard... building a distributed team, even more so. At Elastic — the company behind the open source tools Elasticsearch, Kibana, Beats, and Logstash — everything is distributed; the company and our products. This talk dives into the details of how we are thriving on our distributed model:

• How we started to be distributed by design.
• What our shared values are and what we have learned as we progressed and grew to our current size.
• What the day of an Elastician can look like.
• What we are using in terms of tooling.
• How we hire and onboard new employees.

Did you know there’s a widely-supported browser API that allows you to interface with a wide variety of hardware, including some you can build yourself? Wonder no more and meet the WebMIDI API! We’ll look at how MIDI — a niche protocol from 1983 designed for electronic musical instruments to talk to one another — JavaScript and “tiny computers” like the Arduino, Espruino and RasperryPi can be used to build interactive, non-musical experiences that straddle the line between the web and physical world.

Scaling up your development is never easy task, It's only matter of time until you experience this problem. But it's never the first problem, the first problem is get to market, prove the project/product is actually valuable, fast, efficient and on budget. It's going to be almost impossible to pay the price of scaling from day one.

I will demonstrate how to start a micro-frontends solution even if you working alone. Then I will show how you can scale this solution for multiple teams. Fast & comprehensive kick-start: From nothing to full solution with release pipeline, Angular, React, VUE...

• I assume you already familiar with micro-frontends I will focus on "how" not on "what" and "why"

Modern development processes define code review as a mandatory step before deployment to production. Since the process is done by humans, it is expensive, subjective and requires proper training.

My talk is divided into three parts: first, why code reviews are necessary for ensuring code quality and are critical to the engineering team’s growth. Second, I’ll elaborate on how to conduct focused and efficient code reviews. Finally, I’ll discuss issues that we experienced in our own code review process and explain how we overcame them.

The world of CI tends to be neglected, causing engineering problems both in moral and quality. Things either take too much time, or are quick but unpredictable. In this talk I'll share how we at BigPanda cut down our CI/CD time by drastically, without compromising our confidence in quality, by simply using Git in a more sane way!

You’ve heard a ton of jokes about micromanaging and you get it, really. Yet you have that sinking feeling that if you keep your eyes off the team for a day, things will go haywire. People will stop thinking, somehow you’ll have 3 new programming languages in production, and someone will agree to a deadline that will require acquiring Amazon.

Is this really why you got into management? Is running from meeting to meeting and constantly slack’ing your whole team truly the best manager you can aspire to be?

Let’s talk about creating a team that’s like a good iRobot: set and forget.

Measuring unique users in a billion user network is hard - accurate counting is space consuming and not easily distributable.

In this talk I will describe HyperLogLog, a probabilistic cardinality estimation algorithm and data structure and how we used it to provide breakdowns of our billion user reach.

Disneyland is not just a theme park, it's a magical experience where you can actually live the fairy tale. Where else in the world you can fight Darth Vader, fly together with Peter Pan and hug Mickey Mouse in the same day?

And it’s no surprise, as Disneyland is meticulously and deliberately engineered to be fun & immersive. This practice is called Imagineering.

What if you could use the same principles to bring a magical experience to your users? In this short talk, I’ll share 3 Imagineering principles I’ve been using, and you can use too!

So join me if you want to learn how Amazon.com’s homepage is similar to Cinderella’s Castle, and how snow white helps Airbnb create a better mobile app.

At Cloudinary we use AWS Aurora managed database service that reduces overhead allowing us focus on product development. Aurora comes with a lot of great features out of the box such as: automatic failover in case of master db issue. It also constantly upgrades with new feature, security and performance improvements.

However, upgrade requires up to 1 minute downtime which for heavy load SaaS company is not an option.

In this session, I will present our journey. From the research for proper tools to accomplish zero downtime upgrade with ProxySQL - proven to be a great tool for various scenarios. I will also describe the long POC and issues we faced and will provide code snippets and configuration examples, along with a LIVE demo of Migrating Aurora cluster without any downtime.

Discussing the tools we use as software developers is a normal part of software developer culture. It is normal to dissect the inner workings of a debugger, a compiler or the framework of choice of the moment for the purpose of learning to get more from it. This session is devoted to doing just that to the one tool we ALL use, regardless of programming language or environment but hardly give it any thought - our own mind.

We will discuss the cognitive process of debugging, turning debugging unto itself and gain a glimpse of what we do mentally, emotionally and cognitively when we debug. In the process we will learn how to be better debuggers, programmers and maybe, better human beings.

You may have heard about big companies allowing their engineers to work on whatever they want. You probably thought to yourself "well, they have too much money to waste". Here is where you might be wrong. Hackathons not only improve a company’s product with new innovations, but also provide teams with the opportunity to learn, grow, and share experiences with their colleagues. Haven’t you always wanted to make your [technical] dreams come true?

Companies of all shapes and sizes have recently started to embrace Hackathons. Through this lecture, I will try to list all the right reasons why your company, should be next to hop on that “Hackatrain”. If you already have hackathons at your company, that’s great! The lecture will also share more tips and tricks to help you make them better.

According to AI experts, the art and design scene has been taken by an AI storm, if you’d ask artists though, AI art is just a gimmick. But what do the techies think?

In our 30 minute talk we will dive into the depths of programming a genetic algorithm that plays god and creates forests and other eco-systems on city walls. We will discuss current ML algorithms that form amazing experiences, how do they relate to the history of computer science, some topics in human perception and finally, why and how everyone should create at least one interactive program.

It was late 2017 when we realised that we have to ask our biggest client to stop sending us clickstream events.

We knew something had to change, but we never imagined it would be a full rewrite with a brand new architecture.

Do you have a component that is reaching the boundaries of its ability to scale? So did we! Join me for a tale of how we moved from batch to stream processing, how we went from unstable to solid and scalable technologies, how we reduced latency and increased throughput, all while reducing our cost.

Data-driven decisions is a hot buzzword but it's also a way of life. All of us (product, engineers, analysts, etc) encounter and analyze a lot of information daily and we are expected to use this information to reach conclusions that will affect our products. I will talk about how to approach data analysis, discuss common mistakes made in the process (even by experienced data veterans), share examples from my own experience and provide tips and guidelines for analysis and visualization.

Can you expect your engineers to act as managers? Is it too high of an expectation or will it help them grow? How will it impact your guidance and ability to closely work with them?

In this talk, I will share my experience, insights and conclusions as a manager of managers on this matter.

npm/bigband is an open source tool coming out of testim.IO allowing developers to easily assemble multiple AWS resources (Lambda functions, DynamoDB tables, Kinesis streams) into fully functioning services. Main features include:

• a very simple and intuitive command interface

• automatically decides what needs to be deployed

• Manages and updates IAM policies making them essentially hands-free

• Allows reuse of service definitions from other NPM packages

This tool powers all of Testim's Lambda-based applications. It is now offered to the entire open source community.

After running developer communities over the past 5 years, meeting a lot of great people and learning a lot of things about the Israeli Hi-Tech ecosystem, I’ll share all that I’ve learned about what it actually means to create what is called a “Community”. The value that you as a community lead can give to the people in it and the things that you can gain as a geek out of it. You’ll be surprised to learn how easy and hard it can be at the same time.
I’ll tell about the steps that I’ve taken in this journey, what in my opinion might kill the concept of “Developer Community”, by the end of the talk you'll have as many tools as possible for you to be able to create a community of your own.

Time estimations are an essential part of the software development process. The rise of agile methodologies has even made it easier to come up with more accurate time estimation for engineering tasks. This talk will highlight subtleties, hidden trap doors, and even paradoxes, that await teams that try to over do time estimations. These include:

• time-to-market vs. calendar deadline
• engineering estimation vs. end-to-end estimation
• why is striving to make fully accurate time estimations dangerous to your business?
• are there project types for which time estimations are not as important?
• what is the "black market of time"?

While recruiting a candidate with the required skills is a challenge, their personality should be your main concern. If their technical skills fits, but you still have your doubts, then maybe you haven't examined their emotional intelligence enough?

Well you cant just ask them! All the "tell me about yourself" questions will never reveal their true identity. You must be discreet. You must be devious.

This lecture is about finding the right balance in your expanding team dynamics - analyzing current's team atmosphere and identifying problematic characteristics the candidate might have. After all, we all want to have a perfect match - let me help you with that.

The motivation for a shared UI components catalog is clear - sharing components and resources, leads to a higher product quality, with significantly lower effort. However, this kind of an effort requires workflow discipline, which usually fails. So, how do you keep UI Components implementation aligned with designs? Automated visual tests are critical to maintain a UI components catalog, the last standpoint which makes sure that nothing changes unintentionally. In this lecture I'll share insights and experience regarding visual tests automation and integrating with a UI/UX designer team, which ensures high quality with minimum effort.

The eternal question many hiring managers deal with each day..

What is worth more in today’s job market?

• Credentials (a.k.a. Certification or a degree) or
• Experience (you know your stuff - because you have been around the block - more than once)

I think that most of would agree that the second option has so much more value. This is a question I sometimes ask myself - because I am a 6x AWS certification holder (all in the last 18 months)

This talk will go into the benefits of how pursuing a certification path can give the best of both worlds, both the experience and the accreditation that comes with it.

A win-win situation

And of course why we need to continuously improve ourselves to stay relevant in such a fast-paced technological world.

Designing the right architecture when building a software application is imperative. As serverless based applications are new, it is essential to learn from other team experience as you are adopting the methodology.

In this talk we will discuss:

• Major design and architecture considerations
• Best practices for serverless architecture
• Mono-repo vs. multi-repo
• Fat lambda vs. thin lambdas
• Actionable recommendations for cold starts, scaling, concurrency, layers and more

"I'm quite good in Java, why should I be a Kotlin newbie?"

When we become proficient in the language we have been writing in for years, getting really comfortable with its technological stack, we are afraid to lose all that we have gained and start all over again with some other programming language. For me, this was the case after I gained a lot of experience with C#.

A few years ago I decided to leave my comfort zone and started writing in Scala. Later I added more languages to my toolbox like JavaScript and Python.

I’ve discovered that learning and becoming proficient in more languages has many advantages and benefits and the the continuous learning process can be quite easy.

Over a period of 6 years I’ve worked with 18 different companies. Ranging from tiny startups, through Fortune 100 companies to government units, in 3 continents. You would be surprised how much all of these have in common. Here’s what I’ve seen to work again and again and what I understand never works.

Scala 3 is a major breakthrough for this teenage language. From having too many options in the past, Scala 3 tries to be more opinionated and has gathered many best practices into the language itself.

These include constraining the abuse of implicit conversions, reducing boilerplate for Type Classes and the Tagless Final pattern using implicit function types and a whole new design for enums that enable Algebraic Data Types.

In this talk we will deep dive into these important features that will make our functional code more comprehensible and clean.

Two years ago I quit my job. That’s not really unique, is it? What makes it unusual is that I took this decision less than 6 months after I started. As a veteran R&D manager I know how many efforts are spent on talent acquisition. So I started on a self retrospective move to investigate why did I choose to leave so fast. What went wrong with me?

A month later I started another job. This time I was very aware to what is happening with me as a new employee. I could compare the two onboarding experiences. The lessons I've learned surprised me so much.

In this session, I shed some light on the psychology of a new employee and give practical ways to deal with it. We’ll See through a new employee’s eyes and understand what the ingredients are that can turn onboarding into a success.

Why do engineers feel so miserable when it comes to writing their tests? In this talk, we’ll explore this phenomenon, what is the mentality that’s causing it, truthfully look into the root causes (even if they are a bit embarrassing), and then talk about simple things that can be done to change that mindset. Let’s make tests great again.

Feature flags and user configuration systems are important! They allow PMs and developers to fully control their product behavior, run experiments, tweak user experience and handle a crisis.

When we wanted to enhance our product delivery process at Soluto, we found out that publicly available feature management platforms are in short supply - most companies invest lots of engineering power for building their own in-house solutions.

In this session, I’ll showcase and demo Tweek - our open source, rich, user-friendly feature manager that can be easily used by everyone, anywhere!

So you’ve launched your first service and all your customers in the US love it - great! Now comes the hard part - how do you make the entire world love it too.

In this talk, I’m going to share what from my experience building the PayPal Mobile app and expanding WeWork’s systems to span the globe.
I’m going to cover it all: from encodings, translations, UGC, CATs, i18n, g11n, l10n, LQA, money, dates, names, and taxes. So that you can expand your app and reach new customers in London, Beijing, Tel-Aviv, and Paris.

Hyperledger Fabric is a blockchain framework implementation and one of the Hyperledger projects hosted by The Linux Foundation. In this talk you will learn about Hyperledger Fabric components, such as endorsing peers , ordering nodes , committing peers , consensus , membership services, smart contracts called “chaincode” and more. I will also describe how to build your first Blockchain app using a "code pattern" : creating an asset-transfer app that enables you to transfer marbles between multiple owners

At Soluto we try things to improve our work process, we keep what works, we found this adjustment to work, think it's worth sharing.

Quite often teams find themselves in these situations:

1. A lot of tasks in the Doing phase.
2. Tasks not being completed. Dragging out.....
3. Knowledge Silos - “I don’t know. Talk with X he worked on it.”

A simple concept - Limit work in process --> Don't open new threads! Close exiting ones. Lead to to a small adjustment we did to the way we work which increased our team's focus and shipping speed.

This small change improved the way team members communicate and own tasks, it resulted in improvement to shipping speed, team spirit, and focus.

In this 5 minute talk I’ll show how teams can easily use this.

So you need to move some data from here to there in some given time - easy! You write a few scripts and let *nix Cron handle the tasks. Then you realize that you need a more robust solution with some dependency between tasks and more visualization on them, so you build a few pipelines in Jenkins to handle that. As the data and business needs grow, handling scheduled tasks for different missions at scale becomes too complex:

• Task dependencies (fan-in / fan-out)
• Task priorities & SLA
• Task modularity
• Multiple data sources
• Parallelize tasks
• Visualization of tasks over time
• Task logging
• Task timeouts
• Retries on failure
• Notifications on task error

In this talk we will take a deep dive into Apache Airflow, and how it helps us solve that complexity.

Leaders, either managers or just people who want to make an impact, usually have core values and work culture, that they expect to see in others as well. For example, we’d like our teammates to have a can-do attitude, to think like the customer or to have end to end ownership, but we often struggle with how to make them follow these values. Sometimes, the solution is much simpler than what one would imagine. Sometimes it’s about the words we use and how we use them. A small change that can drive a large impact. In this talk I’ll present a simple, effective way to use words to change your team’s culture.

Building a good services architecture is hard. Building a good services architecture aligned to an evolving business and a massive r&d growth is even a greater challenge. As a R&D Director at Taboola, a super-growing company, I kept asking myself how to overcome this challenge. This is when I came across DDD and Event Storming and became passionate about it.

Domain-Driven-Design, aka DDD, is a well known approach used by Netfix etc for designing complex software with alignment to the business. Event Storming is a tool for implementing this approach, mainly used in EU&US and not IL, at least till I give my talk:-)

In the talk I will explain how to visualize business flows using Event Storming,and how this visualization helps creating biz domain ownership in your engineering organization.

“Just leave the garbage outside and we will take care of it for you”. This is the panacea promised by garbage collection mechanisms built into most software stacks available today. So, we don’t need to think about it anymore, right? Wrong! When misused, garbage collectors can fail miserably. When this happens they slow down your application and lead to unacceptable pauses. In this talk we will go over different garbage collectors approaches in different software runtimes and what are the conditions which enable them to function well.

https://github.com/dmarcous/S2Utilities This projects aims at providing a set of tools to make working with geospatial objects quick and painless. These tools were designed with S2 objects (Google's "geometry on a sphere" abstractions) in mind as the leading data structure to be used when working with geospatial data.

• Written in Scala

In 2015, compromised developers unwittingly used a malicious XCode to inject malware into thousands of apps, stealing data of millions of users. In both 2014 and 2017, millions of Uber user and driver records were leaked following an improperly secured token commited to a GitHub repo.

What do all of these have in common? We, developers, caused them.

In this talk, we’ll better understand the damage we can cause, from distributing trojans to exposing user code. We’ll better understand the techniques attackers use to compromise us, so we can be prepared. Most importantly, we’ll talk about how we can defend ourselves – and our users.

Analytic data, heat maps and screen recordings are the bread and butter of improving the user experience. By using these tools, we can literally see and understand our users’ behavior and from there, translate these insights into an even better experience.

But do these tools help us understand and solve technical problems? Here at Soluto, yes! They help us in understanding our users’ experience from a technical perspective.

In this session, I’d like to share how common tools like screen recordings help us troubleshoot through our users’ eyes. This process gives us invaluable data and helps us gain empathy for our users in order to turn our products from good to exceptional.

Node.js has been around for a while now, and most will consider it stable. But what happens when you turn your head away for a month or twelve? In this talk, I will share the adventures we went through to update a simple node.js app after not touching it for a year. It's a story about the good, the bad, and the broken API's.

Any CS student can tell you that abstractions are good, and tech debt is bad. that code duplication is forbidden, and supporting future feature is important. Is that true?

In this talk, I will challenge some of the well-known programming conceptions and principals, in order to help you be more skeptical and critical regarding popular perceptions and to write better code.

YOU CAN NEVER KNOW WHERE YOUR NEXT GREAT IDEA COMES FROM!

We were struggling with a big challenge and after evaluating a variety of approaches it felt like a dead end. And then we found the answer! It was at the 2017 Reversim summit following a session about a thirty years old algorithm mimicking a flock of birds. It was an aha moment for us - taking us to an almost two years path of solving the challenge by applying biomimicking algorithms.

I will present the concept of biomimicry and treating biological processes as algorithms nature has created. I will also take you through a specific use case and demonstrate how treating job postings as a flock of birds can optimize budget spend!

By the end of the session, I hope you have a different perspective on both nature and machine learning.

DevOps is a world surrounded by information, starting from a single commit and ending in roll out to production. In this talk, I'll introduce you to the world of Taboola DevOps data collection, to better understand what goes on under the hood. The system we've developed in-house helps us collect and analyse the entire DevOps process from the very first commit all the way to production. It provides us a full clear view with a drill-down toolset that helps keep us away from the dark side. Our KPI's moved from being abstracted ideas to data driven goals, which we can measure and act upon.

What does it mean to strive for ENGINEERING EXCELLENCE? As your engineering organization grows, you want to encourage teams and individuals to develop and improve their technical proficiency and practices. Growth puts stress on your teams' ability to deliver kick-ass code consistently. To help them improve and grow, we built the LEVEL UP model to represent everything that's expected of an excellent engineering team.

In this talk, I’ll review why we built this model, refined it and how we got teams to adopt it and the effects that it had on our department. I will also give you the tools to either use our model or fork your own specific one for your organization.

Transferring files between nodes is a common task. When these files are large it also consumes time. Time we usually don't have. A very commonly used utility is 'scp'. This is the story of how I worked on a task and as a by-product reduced transfer time by 70%. And now it is available as an open-source for all.

During the past 8 years, Fiverr migrated from a LAMP architecture all the way to 200~ Microservices with more than a dozen teams pushing a few hundred deploys a week.

Working inside the monolith was simple, all data was available but the move to microservices presented challenges as data was exposed via REST API, which caused performance and reliability issues, dependencies between teams and much more...

If you feel that your current strategy for sharing data is brittle this talk is for you! I'll present our robust solution for sharing data across Microservices, utilizing Kafka as a data hub using techniques borrowed from CQRS and DDD to tie everything together and iterate much faster!

I never imagined I’d know who Indian cricket star Rohit Sharma is, but then traffic to our real-time HTTP request sender service suddenly increased from 20 to 40 million events per minute. The reason? An app streaming the first game of the Indian cricket season. While growth is a good thing, we found ourselves unprepared for this sudden spike and we needed to scramble. Initially, we just threw money at the problem, but this wasn't sustainable. My talk will describe how we found low-cost, programmatic and architectural solutions to this problem and how we prepared ourselves to handle massive spikes like these on top of our existing 70 billion events per day. I'll explain our process of profiling, performance enhancement techniques, and some important lessons learned along the way.

In this talk we will review the option out there to solve how do you manage your repo(s) in the microservices world. Would you go with multi repo? mono repo? what tools are out there to help you out? we will also review the difference between services and infrastructure packages This talk will focus on NodeJS and NPM

I’m a huge fan of game jams which force you to create something fun and playable in a short period of time. For years now I’ve used tools like Sfxr which generates 8-bit sound effects based on predefined templates. When I started developing my fantasy game console, Pixel Vision 8, I wanted to see if I could build upon Sfxr and create randomly generated music for those of us that know nothing about musical composition. To do this, I employed a collection of techniques to help teach my software how to make “good” music randomly based on a set of predefined musical styles.

In this talk, I’ll walk you through how PV8’s music generator works, what it’s capable of doing, and how I plan to leverage machine learning to help take it to the next level.

While there are many paradigms for scaling dev teams, this is not the case for data science teams.

In an attempt to keep up with company goals and growing demand for improved classification models, we had to shorten the research life cycle, and release the bottle neck of training, evaluating and deploying classification models. In this talk I’ll present the design of our model training platform - a set of independent tools and libraries, which allowed us to drastically increase the number of experiments we can perform, apply our findings easily to all existing models, and standardize the evaluation process.

Adopted by Google in 1999, objectives and key results are a simple way to align, focus and engage a team around company goals.

In this talk, I will show you how we are using OKRs at Nanit from the company's vision down to the sprint tasks.

I will also talk about adopting OKRs for personal goals

Optimizing auction revenue from the seller's perspective is a well studied problem in Game Theory. However, as is often the case, practical auctions extend well beyond the simplifying assumptions made in the literature. This talk will exemplify some of the gaps between theory and the optimization of real-time bidding ad auctions at Outbrain, and will highlight areas where theoretical ideas may still be productized despite the gaps.

“Why on earth did we decide to develop our own RPC system?” is a simple and obvious question to ask, but a complex one to answer. In this talk, Dr. Robert Fink, Head of R&D at Palantir Technologies, will give a firsthand account of the development history of Palantir's polyglot RPC system, Conjure, from the first line of code to its rebirth in open-source. We discuss the rationale and design principles behind Conjure, its implementation, and the lessons learnt on the way (“self-hosting intermediate representation”, anyone?). If you are looking for a simple yet strongly-typed RPC system that works for servers and clients in Java, Go, Rust, Python, and TypeScript, maybe look no further than Conjure?

Data Science teams are usually highly-skilled and enthusiastic, but do they manage to stay relevant in face of extreme growth and an intense Agile environment?

In this talk, I'll suggest that although intuitively it makes sense to have an unstructured Data Science team that's not bound to agile principles, without any concern for execution and delivery - ultimately this is counter-productive. I will review some fundamental steps to structure an impactful Data Science team, which includes creating a proper environment and work culture.

In my experience as a Data Science leader, once we’ve transitioned to become an Agile team that is focused on clear business KPIs, we’ve managed to become a strong and independent force, significantly impacting the company’s success.

Have you heard about Vue.js but not really sure what it is or why it's so popular? This one is for you! In my talk, we will go over some key features that make Vue.js stand out in the framework crowd, explore its advantages and discover how easy it is to build scalable applications with Vue.js