The Elephant in the Room: Feature Engineering

Description

We pretend it’s easy. Fifty tables, hundreds of columns, and somehow you’re supposed to know exactly which variable determines if a transaction is fraud. You spend days browsing schemas, running describe queries, guessing at joins — and this is before you’ve written a single line of ML code. Feature engineering and data understanding eat up 70% of the ML lifecycle, yet we keep treating it like a solved problem. Meanwhile, frontier models could help — but we don’t trust them with our data, and rightly so.

What if the model never saw your data at all? We built a tool that points a local LLM (Ollama, fully offline, open weight models) at your databases, discovers schemas, computes statistics using native SQL, and runs EDA — all without a single row leaving your machine. Then it suggests production-grade ML features — entity keys, transformations, types — and you review them in a table. Keep what makes sense, throw out what doesn’t. One click generates a complete Feast feature store repo. Another click runs feast apply. Schema to deployed feature store in under a minute. No API keys, no cloud, no trust compromise. Open source models solving the problem that actually takes up your time.

Key Takeaways

1. Feature engineering is the real bottleneck, not modeling — and open weight models are good enough to help, if you design the right guardrails. The LLM never touches raw data. It only sees schema metadata and pre-computed statistics. You review every suggestion before anything gets applied.
2. The pattern generalizes beyond Feast — “LLM suggests, system validates, human decides” works for any ML platform. We’ll show real outputs from 3B and 8B models, where structured prompting works, and where small models break down.

Target Audience

ML engineers, data scientists, and platform engineers who spend more time wrangling data than building models — and are curious about using open weight models to fix that, without sending data to the cloud.

Bio

Aniket Paluskar — Curious Engineer at Red Hat AI working on Feast & Data.
Chaitany Patel — Associate Software Engineer at Red Hat working on Feast (#1 open source feature store).

Github

https://github.com/patelchaitany/schema-discovery-eda