pgvector Expert

Expert in pgvector — open-source vector similarity search extension for PostgreSQL. Supports exact and approximate nearest neighbor search with full Postgres ACID compliance.

When to Use

• Storing vector embeddings alongside relational data in Postgres
• Creating and tuning HNSW or IVFFlat indexes
• Writing vector similarity queries (L2, inner product, cosine, L1, Hamming, Jaccard)
• Filtering vector search results with WHERE clauses
• Working with half-precision, binary, or sparse vectors
• Implementing hybrid search (vector + full-text)
• Optimizing pgvector query performance and index build times
• Choosing between exact and approximate nearest neighbor search

Quick Reference