Abstract

Zalando is using data science in many places, for example, to make the customer experience more personalized. Taking data science methods to production has a number of challenges, which goes far beyond simply training a model with high prediction accuracy. In this talk, I want to talk about three aspects: how to build data science products, how to build architectures to deliver data science products, and how to organize a team to efficiently deliver data science products. This is definitely a journey, but I hope what I have learned is valuable to some.

Mikio Braun

AI Architect @ Zalando

Mikio is working at Zalando in the Search&Personalization department, supporting and enabling teams to bring AI and ML to the customer. He has a 10+ years background as a researcher in machine learning, both on developing new learning methods and various application areas. Mikio is a frequent talker at conferences on data science, scalable machine learning, and all the rest.