Skip to content

Volume 9 — Retrieval-Augmented Generation

Grounding models in your data

Embeddings, vector stores, chunking, hybrid search, reranking and the evaluation that tells you whether your RAG system is actually better than the baseline.

Outlined, not yet written

This volume is fully outlined but the lessons are not yet authored. The chapter list below is the plan of record, not a promise that content exists.

What you will be able to do

  • Build a RAG pipeline end to end
  • Measure retrieval quality separately from generation quality
  • Diagnose whether a bad answer is a retrieval or a generation failure
  • Justify chunk size and overlap with evidence rather than folklore

Chapters

  1. Why RAG exists
  2. Embeddings and vector similarity
  3. Chunking strategies
  4. Vector databases and indexes
  5. Hybrid search: dense plus sparse
  6. Reranking
  7. Query rewriting
  8. Citations and grounding
  9. Evaluating retrieval quality
  10. End-to-end RAG evaluation
  11. RAG failure modes

All volumes