AI & DataIntermediate
Vector Database
PronunciationVEK-ter DAY-ta-base
Definition
A specialized database designed to store, manage, and query high-dimensional vector embeddings efficiently using similarity search.
Where you hear it
In AI and machine learning discussions, particularly when building RAG systems or semantic search engines.
Examples
We need a vector database to store the document embeddings for our semantic search.
The new vector database allows us to perform similarity searches across millions of items in milliseconds.
Common mistake
Thinking a traditional relational database can handle high-dimensional vector similarity search with the same speed and indexing techniques as a dedicated vector database.