Vector DB Expert · mlun6440-bit bundle Expert guidance on vector embeddings, vector databases, and retrieval-augmented generation (RAG) systems. Use when users ask about choosing embedding models (OpenAI, Voyage AI, Cohere, open-source), vector database selection (pgvector/Supabase, Pinecone, Qdrant, Weaviate, Milvus/Zilliz, Chroma), chunking strategies for documents, HNSW or IVFFlat index tuning, hybrid search (BM25 + vector), re-ranking with cross-encoders, Matryoshka/MRL dimension reduction, embedding quantization, RAG pipeline architecture, contextual retrieval, GraphRAG vs traditional RAG, pgvector performance tuning, Supabase vector search functions, similarity metrics (cosine, dot product, L2), metadata filtering, multi-tenancy with RLS, production scaling, cost optimization, or any question about storing, indexing, and searching vector embeddings. Also use when building or debugging RAG applications, designing semantic search systems, or migrating between vector databases.
