Scalable Feature Engineering with Hamilton on Ray

Discover how to scale feature engineering using Hamilton on Ray at StitchFix in this informative 30-minute conference talk. Learn about Hamilton, an open-source, declarative, general-purpose dataflow micro-framework written in Python, originally created to manage the complexities of scaling a team and time-series feature engineering codebase at Stitch Fix. Explore what Hamilton is, why it was developed, and how to leverage it for feature engineering. Gain insights into scaling computation effortlessly with the out-of-the-box Ray integration. Through code examples in slides and a quick demo, understand how a Data Science team at Stitch Fix scaled their team and codebase using Hamilton, grasp the declarative API paradigm it prescribes compared to traditional approaches, and learn how to utilize the Ray integration with Hamilton for enhanced scalability.

Scalable feature engineering with Hamilton on Ray at StitchFix