AI Terminology Course
AI Terminology
/
Advanced

Vector Database

Definition

A specialized database designed to store, manage, and search high-dimensional mathematical vectors (embeddings) instead of traditional rows and columns.

Explain Like I'm New

A map of concepts. Instead of an Excel sheet, it's a 3D universe where similar ideas float near each other. You throw a new question into the universe, and the database instantly returns the 5 documents floating closest to it.

Real World Example

Pinecone, Weaviate, or ChromaDB. When you upload 1,000 PDFs to a RAG system, they are converted into vectors and stored in a Vector Database for lightning-fast semantic search.

Common Use Cases

  • •RAG infrastructure
  • •Recommendation engines

Interview Questions

basic

  • Can a traditional SQL database easily perform fast semantic similarity searches?

intermediate

  • What mathematical operation do Vector Databases use to find the 'closest' document?

Flash Cards

Question

Can SQL do it?

Click to reveal answer
Answer

Historically, no. SQL is built for exact text matches and numbers. (However, modern SQL databases like PostgreSQL have added plugins like `pgvector` to support it).

Question

Which math operation?

Click to reveal answer
Answer

Cosine Similarity (or sometimes Euclidean Distance). It calculates the mathematical angle between two vectors to determine how similar they are.