AI
Vector Database
A database optimised for storing embeddings and finding the nearest matches to a query vector at high speed.
Where a normal database is good at exact matches ('find the row where id = 42'), a vector database is good at similarity searches ('find the ten passages most similar to this question'). It's the storage and retrieval engine behind every RAG system.
From a product-team perspective the vector DB is plumbing. What matters is what you index in it: a per-product corpus of structured manuals, FAQs and specs will outperform a generic dump of the company website every time.