Vector Database

Definitions

Vector Database is a specialized database system optimized for storing, indexing, and querying high-dimensional vectors (arrays of numbers) that represent data points in a continuous mathematical space. In the context of AI and machine learning, these vectors are typically embeddings, dense numerical representations of text, images, audio, or other data generated by neural networks, where semantically similar items are positioned close together in the vector space. Vector databases enable similarity search: given a query vector, find the nearest neighbors efficiently using algorithms like HNSW (Hierarchical Navigable Small World), IVF (Inverted File Index), or product quantization. This capability powers semantic search (finding documents by meaning rather than keywords), recommendation systems, image similarity search, RAG (Retrieval-Augmented Generation) for LLMs, and anomaly detection. Popular vector databases include Pinecone, Weaviate, Milvus, Qdrant, and Chroma. Traditional databases like PostgreSQL (with pgvector) and Elasticsearch have added vector search capabilities. The rise of LLM applications has driven explosive growth in this database category.
In plain English: A special database for finding things by meaning rather than exact words — it stores numerical representations of content and finds the closest matches.
Example: "Store the document embeddings in Pinecone so we can do semantic search instead of keyword matching."

2017: Facebook AI Research released FAISS (Facebook AI Similarity Search), a library for efficient similarity search on dense vectors, addressing a need in recommendation and retrieval systems.
2019: Purpose-built vector databases like Milvus and Pinecone emerged, optimized for storing, indexing, and querying high-dimensional embedding vectors at scale.
2022: The generative AI boom drove massive interest in vector databases for retrieval-augmented generation (RAG). Weaviate, Qdrant, and Chroma gained rapid adoption.
2023-Present: Traditional databases (PostgreSQL via pgvector, Elasticsearch, Redis) added vector search capabilities. Vector databases became essential infrastructure for AI-powered applications.