Vector Database
Also: vector store
A database built specifically for storing and searching AI-generated numerical representations of text (called embeddings) — not traditional rows and columns. When you build an AI system that can answer questions from a large document collection, you need a way to quickly find the most relevant documents for any query. Vector databases do this efficiently by measuring how 'close' query vectors are to document vectors. Common examples: Pinecone, Weaviate, Qdrant, and pgvector (a PostgreSQL extension).
In practice
You want Claude to answer questions from 10,000 internal documents. A vector database stores each document chunk as a set of numbers (an embedding) that captures its meaning. When a question comes in, the database finds the chunks with the most similar meaning and passes them to Claude. Pinecone, Weaviate, and pgvector are common choices.
Related concepts