AI Infrastructure Glossary

Vector Database

A database that stores information as mathematical vector representations (embeddings), enabling semantic search by meaning rather than exact keyword matching.

Also known as: vector store, embedding database

A vector database stores data as high-dimensional numerical vectors called embeddings. Unlike traditional databases that retrieve records by exact matches, vector databases retrieve records by semantic similarity ... finding content that is meaningfully similar to a query even when the words do not match. AI search systems use vector databases to index webpage content as embeddings, then retrieve the most semantically relevant passages when answering questions. This is why clear, topic-focused writing improves AI search performance: the semantic meaning of your content directly affects how it is indexed and retrieved.

Related Terms

  • RAG (Retrieval-Augmented Generation) — A technique where an AI system retrieves relevant information from external sources before generating a response, grounding its output in real data rather than model memory alone.