Glossary
Vector database
A store for embeddings that answers 'what is semantically closest to this?' quickly.
Vector databases index high-dimensional embeddings and retrieve nearest neighbours efficiently. They are the retrieval substrate under most RAG and semantic search systems.
You often do not need a separate one. pgvector puts vector search inside Postgres, keeping relational and vector data in one system with transactions across both.
Choosing one is mostly a scale question. Below a few million vectors, pgvector inside your existing Postgres is almost always the right answer and saves you an entire system to operate. Above that, or where you need very low latency at high query volume, a purpose-built index starts earning its keep. Migrating later is unpleasant but tractable; adopting one too early costs you every day until then.
Commonly misunderstood: A dedicated vector database is frequently adopted before the workload needs it, adding a second system and a second bill for no measured gain.
Related terms, in context
The concepts you almost always meet alongside vector database.
- Embedding
- A numeric vector representing text, images or audio, positioned so similar things sit close together.
- RAG
- Retrieving relevant passages from your own documents and giving them to the model, so answers are grounded and citable.
- Hybrid search
- Combining keyword and vector retrieval, because each fails where the other succeeds.
Where this shows up in our work
Vector database is not an abstraction for us. It is a decision we make on live projects. It shows up most directly in rag & knowledge retrieval, ai search implementation, where getting it wrong has a cost someone can measure.
If you are evaluating a vendor on this, the useful question is not whether they can define the term. It is what they measure, what they would refuse to do, and what happens in their system when the assumption behind vector database stops holding.
Questions
What is Vector database?
A store for embeddings that answers 'what is semantically closest to this?' quickly.
What do people get wrong about vector database?
A dedicated vector database is frequently adopted before the workload needs it, adding a second system and a second bill for no measured gain.
Does Orqent Labs build this?
Yes, RAG & Knowledge Retrieval Systems and AI Search Implementation. We work across India, covering all 19,238 PIN codes remotely.
Building something that involves vector database?
We will tell you honestly whether it is the right approach for your problem.
Or email bd@dtrasglobal.com · call +91 74118 77878