watch

Videos

Free overviews on applied generative AI. The unlisted deep-dive teardowns live in the premium portal.

Evals that predict production, not vibes

4:12

A free overview of why most LLM eval suites pass while production regresses — and the offline/online split that actually catches it.

RAG is becoming retrieval-plus-reasoning

6:48

Why naive vector RAG underperforms, and the hybrid-search + re-ranking stack that fixes most of it.