What Is a Vector Database?
A vector database stores embeddings so you can search by meaning, not keywords. It's the retrieval engine behind RAG chatbots and semantic search.
Vector database: a database that stores embeddings (numeric representations of meaning) so you can search by similarity: find the passages, products or tickets closest in meaning to a query, not just those sharing keywords.
Vector database is a database that stores embeddings (numeric representations of meaning) so you can search by similarity: find the passages, products or tickets closest in meaning to a query, not just those sharing keywords.
Text, images or records are converted to vectors by an embedding model; the database indexes them for fast nearest-neighbour search. This is what lets a RAG system fetch the right context, or a search box understand ‘refund policy for damaged goods’ without those exact words appearing.
Why it matters
Semantic search is the backbone of most useful GenAI features. The choice of vector store (pgvector inside your existing Postgres, or a managed service like Pinecone or Weaviate) affects cost, ops burden and how it fits your stack.
How Finzarc thinks about it
We usually start with pgvector when you already run Postgres (one less system to operate) and move to a dedicated store only when scale demands it. See talk-to-your-data analytics we’ve built on this pattern.
Related
Want this built into your business, not just explained? See what we ship or book a 30-minute scope call.
Questions, answered.
What is a vector database used for?
It stores embeddings so you can search by meaning, powering RAG chatbots, semantic search, recommendations and deduplication. You give it a query vector and it returns the closest items.
Do I need a dedicated vector database?
Not always. If you already run Postgres, the pgvector extension is often enough and keeps your stack simple. Dedicated stores like Pinecone or Weaviate make sense at large scale or for specialised indexing needs.
What's the difference from a normal database?
A normal database matches exact values and keywords; a vector database matches by similarity in meaning, using nearest-neighbour search over embeddings. Many systems use both together.
30 minutes with the founding team. Bring the problem; leave with a scope, a timeline, and the number it should move.