#pgvector

"Do we need Pinecone for this?" If you're adding AI search to an app in 2026, someone on your team...

What a vector database does, how HNSW and IVFFlat work, and the traps of running pgvector inside the Postgres you already run.

Operator notes, temperatures, and 384-dimension embeddings live on a TimescaleDB hypertable. Cosine search, a site filter, and a time window all run in one SQL statement.

Why I Started Exploring Semantic Search I'm building Second Memory, a personal memory...

Over the past two years a familiar story has played out across the industry: a team reads a benchmark...

How dense vector retrieval paired with sparse BM25 indexing and cross-encoder re-ranking drives 95%+ precision in enterprise knowledge bases.

Why Postgres as the AI Backend The pitch for building AI features directly on Azure...

The short version What is the smallest number of containers you need to evaluate a...

Second-Memory stores things a user writes over time. If a user later asks: “What did I write about...

Build a production-grade RAG app with Supabase pgvector, Next.js App Router, hybrid search, reranking, streaming responses, evals, and cost tracking.

Somebody on your team wants to add search. There are three obvious answers and a lot of confident...

How I built a RAG retrieval service using pgvector, migrated embedding models under pressure, and added provenance tracking for evaluation calibration.