5 of 5 parts published
Befriending Vector Databases
A practical series on vector databases, semantic retrieval, and how to choose the right stack for real-world AI applications.
A practical overview of vector databases, why they matter for AI retrieval, and where leading options shine.
Embeddings turn text, code, and media into vectors you can search by meaning — plus how model choice and chunking quietly make or break retrieval.
A practical guide to similarity search: vectors, distance metrics, ANN indexes, recall/latency tradeoffs, filtering, hybrid search, and reranking.
A hands-on mental model for vector indexing: why brute force fails, how HNSW and IVF work, what PQ compresses, and which knobs matter in real databases.
Finish the series by building a tiny local semantic search tool with Postgres, pgVector, embeddings, and a command-line query loop.