Data Science projects in startups have some unique characteristics when compared to both software development projects and more traditional research projects. Both the data scientist and her peers can find the respective differences unintuitive and confusing, and find some aspects of the typical flow of both lacking when used to run an agile data science project.

The aim of this talk is to present the characteristic project flow that I have identified in the working process of both my colleagues and myself in recent years, working as data scientists in various startups. Hopefully, this can help both data scientists and the people working with them to structure agile data science projects in a way that reflects their uniqueness.