I built a feature store in pure Python to finally understand the point-in-time join

For a long time the phrase "feature store" sat in my head as a box on an architecture diagram. Feast, Tecton, Databricks Feature Store. I knew they served features to models and I knew they were a Real Thing that ML platform teams care about, but I could not have told you what the hard part actually was. So I did the thing that always works for me: I rebuilt a tiny version from scratch, with no pandas and no pyarrow, just the Python standard library and sqlite. The project is called Asof, and building it taught me that the whole category exists to solve a single deceptively nasty problem.

That problem is the point-in-time join, and once it clicked I could not unsee it.

Live demo and system map: https://hajirufai.github.io/asof/

Code: https://github.com/hajirufai/asof

The one problem a feature store is really for

I built a feature store in pure Python to finally understand the point-in-time join

Related reading

Feature Stores and the Problem They Solve

The Feature Store: Consistency and Latency Are Both Non-Negotiable

ML Feature Store Versioning: Reproducible Quants

Introducing Redis Feature Form

How a unified data model improves feature flag rollout decisions | Datadog

Building a B+ Tree Storage Engine (Part 1): Why Order Stops Making Sense on Disk