Vector Databases and Hybrid Search
Choosing and operating the retrieval layer: pgvector vs dedicated vector databases, ANN indexing internals, filtering, hybrid search and re-ranking
A systems-level treatment of the retrieval layer for RAG: pgvector vs dedicated vector databases (Pinecone, Weaviate, Chroma, Milvus) with a comparison table and a 'when Postgres is enough' decision rule; HNSW vs IVF ANN indexing and their recall/latency trade-offs; pre-filter vs post-filter vs filtered-ANN and why naive post-filtering silently starves results; hybrid dense+BM25 search with reciprocal rank fusion (RRF) worked by hand; and cross-encoder re-ranking with its O(k) cost model.
Practice questions (5)
-
View →
pgvector or a Dedicated Vector Database?
Advanced · Free -
View →
HNSW vs IVF: Choosing and Tuning an ANN Index
Advanced -
View →
Diagnosing a Silent Zero-Results Bug
Advanced -
View →
Hybrid Search: Fusing Two Ranked Lists with RRF
Advanced -
View →
Budgeting a Two-Stage Retrieval Pipeline with Re-Ranking
Advanced