AI Infrastructure Glossary
Definitions for vector databases, embeddings, RAG, semantic search, and the technical infrastructure behind AI-powered retrieval and search.
R
- 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.
V
- Vector Database
- A database that stores information as mathematical vector representations (embeddings), enabling semantic search by meaning rather than exact keyword matching.