Vector Database
A database built to store embeddings and find the ones most similar to a query, fast. It's what lets AI apps search by meaning instead of by exact keyword.
Think of It Like This
A librarian who files books by what they're about, not by how their title is spelled.
Text, images, or other items get turned into embedding vectors, and the vector database indexes them so it can return the nearest neighbors in milliseconds — even across millions of entries. That similarity search is the "retrieval" half of retrieval-augmented generation. Under the hood it leans on approximate nearest-neighbor algorithms, trading a sliver of accuracy for the speed to stay responsive at scale.