Scalable feature engineering with Hamilton on Ray at StitchFix

Scalable feature engineering with Hamilton on Ray at StitchFix
Hamilton (github.com/sti...) is an open source, declarative, general purpose, dataflow micro-framework, written in Python. It was originally created to manage complexities of scaling a team along with a time-series feature engineering code base past thousands of features at Stitch Fix. At a high level, in this talk, we'll cover: what Hamilton is and why it was created, how to use it for feature engineering, and how you can scale computation easily with the out-of-the-box Ray integration. At a low level, through code in the slides and a quick demo, you'll walk away with an understanding how a Data Science team at Stitch Fix scaled their team and code base with Hamilton, what Hamilton is and the declarative API paradigm it prescribes as opposed to traditional approaches, and lastly how the Ray integration with Hamilton works and how you can utilize it.
See all Ray Summit content @ anyscale.com/ra...