AI & Web Glossary
Vector database
A vector database stores embeddings — the numeric representations of your text — and finds the closest matches to a query in milliseconds. It is the search index behind document AI and knowledge assistants.
A vector database is a database built for one job: store millions of embeddings and, given a new one, find the closest matches fast. Where a normal database answers 'find rows where customer = Smith,' a vector database answers 'find the passages that mean roughly the same as this question.'
In a typical document AI setup, your files get split into passages, each passage becomes an embedding, and the embeddings go into the vector database. When someone asks a question, the system embeds the question, pulls the nearest passages, and hands them to the language model to compose an answer. The vector database is the card catalog; the model is the librarian reading you the answer.
Practical note for budgeting: this is one of those infrastructure pieces that's invisible in the demo and real in production. It needs hosting, backups, and updating as documents change, modest costs, but actual ones. For most small-business document sets, lightweight options run perfectly well on ordinary servers.
Where this shows up in practice
- Our Knowledge & Document AI service puts this to work for small and mid-sized businesses.
- From the blog: The infrastructure bill nobody mentions when you adopt AI
- From the blog: The unglamorous work of document AI