What you'll be able to do
- ✓Pick and evaluate embedding models against your own corpus
- ✓Combine keyword and vector retrieval into a hybrid pipeline
- ✓Chunk documents without destroying their meaning
- ✓Measure retrieval quality with real metrics, not vibes
Who it's for
Engineers building search or RAG features on top of their own content.
Contents
- 01Embeddings from first principles
- 02Index types and their trade-offs
- 03Chunking and document preparation
- 04Hybrid retrieval and reranking
- 05Evaluation harnesses
- 06Serving retrieval at scale
