What is Vector Database?
Turkish: Vektör Veritabanı
A vector database stores embeddings and retrieves records by semantic similarity, making it a core layer in AI search systems.
What is a Vector Database?
A vector database stores content such as text, images, or product records as numeric embeddings and answers the question “is this similar in meaning?” instead of only “does this exact keyword match?” That makes it useful for finding related documents, equivalent intent, and near matches that classic keyword search can miss.
How It Works
Content is first converted into a high-dimensional vector by an embedding model. When a user searches, the query is embedded with the same model, and the database retrieves the nearest vectors. Similarity is commonly calculated with cosine similarity, dot product, or Euclidean distance.
Pinecone, Weaviate, Qdrant, Milvus, and pgvector on PostgreSQL are common options. Embedding quality depends on the model, while LLM-based RAG systems pass the retrieved document chunks into the answer as context. Teams already using PostgreSQL can often add pgvector before adopting a separate service.
Business Use
Vector databases are used for document search, support bots, product recommendations, candidate-CV matching, and internal knowledge base search. For example, a company assistant can match “late delivery penalty” to contract clauses that use different wording but describe the same issue.
The main design risks are stale indexes, weak access control, embedding cost, and poor retrieval quality. Sensitive documents need permission checks at search time, not only when the data is originally indexed.
Related Terms
Chunking splits long text into meaningful, manageable passages that search and RAG systems can retrieve accurately.
EmbeddingAn embedding represents text, images, products, or other data as numeric vectors that can be compared for semantic similarity.
LLM (Large Language Model)An LLM is a model trained on large text datasets that can understand and generate natural language, forming the basis of tools like ChatGPT.
PostgreSQLPostgreSQL is an open-source ACID-compliant relational database that combines SQL, JSON support, indexing, and extensibility.
RAG (Retrieval-Augmented Generation)RAG is an AI architecture where a language model retrieves relevant passages from documents or databases before generating an answer.
Semantic SearchSemantic search finds relevant results by comparing the meaning of queries and content, not only matching exact keywords.
Vector Index (pgvector)A vector index speeds up similarity search by organizing embeddings so nearest neighbors can be found efficiently at scale.