architecture

Vector database

A vector database stores embeddings efficiently and answers similarity searches in milliseconds — the storage behind every RAG system.

Also known as: Vector DB, Vektor-Datenbank

In detail

While a classic SQL database is optimized for value comparisons, vector DBs are built for nearest-neighbour search in high-dimensional spaces. Common choices:

pgvector (Postgres extension) — our default for most projects
Qdrant and Weaviate — standalone vector DBs with more features
Pinecone — cloud-only, US-hosted (we don't use it)

We prefer pgvector: runs in the same Postgres instance as your business data, simpler backups, EU hosting trivial.

Related terms

Embedding
An embedding is a numeric representation of text (or an image) in a high-dimensional space where similar content sits close together — the foundation of semantic search and RAG.
Multi-agent
Multi-agent refers to an architecture where several specialized AI agents collaborate — a router decides who takes over, and they hand off tasks cleanly.
RAG
RAG (Retrieval-Augmented Generation) is a technique where the language model retrieves relevant documents from your knowledge base before each answer and uses them as the basis for its response — so the agent stays current and answers with sources.