Section · 1 articles

Learn

Teaching material written by practitioners, organised in series.

RAG

Series

Series · 4 parts

Agents without magic

Tool calling, planning loops, memory and failure modes — implemented in 200 lines, then broken on purpose.

Series · 6 parts

RAG from first principles

Chunking, embeddings, hybrid search, reranking, evaluation, and the part nobody talks about: updating the index.

Series · 3 parts

Statistics for people who ship models

Variance, confidence intervals and why your 2-point improvement might be noise.

Learn · RAG

Chunking, explained with a ruler and a pair of scissors

Before embeddings, before vector databases, there is a much dumber question: where do you cut the document? It turns out the answer decides most of your retrieval quality.

S. Dubois · 28 Aug · 3 min