Vector database: A vector database stores embeddings, the numeric representations of text, images, or other content, and finds the items closest in meaning to a query quickly. It is the storage layer behind semantic search and most document-aware AI assistants.
Once your documents are converted to embeddings, you need somewhere to keep millions of these numeric vectors and a fast way to ask 'which ones are nearest to this new vector?' That nearest-neighbor search is what a vector database is built for. A regular database can filter by exact values; a vector database ranks by similarity.
Products in this space include dedicated services such as Pinecone, Weaviate, Qdrant, and Milvus, plus extensions like pgvector that add vector search to a standard Postgres database. Many teams start with the extension because they already run Postgres.
The database is only as good as what you put in it. Chunking documents sensibly, keeping metadata like source, date, and department attached to each vector, and re-indexing when documents change all matter more than which product you pick. Stale vectors return stale answers.
Example at work
A law firm's knowledge team indexes ten years of internal memos and precedent summaries into a vector database with metadata for practice area and date. Attorneys ask questions in plain language, retrieve the closest passages, and get citations back to the original documents, all inside a tool the firm controls for confidentiality.
Why it matters
If your company builds or buys a document-aware AI assistant, a vector database will be part of it. Knowing what it does helps you ask practical questions about what gets indexed, how often it refreshes, and who can see which documents.
Related terms
- EmbeddingAn embedding is a list of numbers that represents the meaning of a piece of text, an image, or other content, so that items with similar meaning end up with similar numbers. Embeddings are what let AI systems search by meaning instead of exact words.
- Semantic searchSemantic search finds results based on the meaning of a query rather than exact keyword matches, so a search for 'unpaid invoices' also returns documents about 'overdue bills.' It uses embeddings to compare meaning.
- Retrieval-augmented generation (RAG)Retrieval-augmented generation (RAG) is a technique in which an AI system first searches a set of documents for relevant passages and then gives those passages to a language model to answer from. It lets an assistant use your own, current information.
- Data privacyData privacy, in the context of AI, is the set of rules and practices that govern what information you put into an AI system, who can see it, how long the vendor keeps it, and whether it is used to train future models.