---
title: "pgvector | Skills We Assign For | Azendo"
description: "pgvector for vector search in Postgres — one system, transactional consistency, plus the Azendo roles assigned for it."
url: "https://azendo.co/skills/pgvector/"
---

[Skills](https://azendo.co/skills/) AI and machine learning 

# pgvector.

pgvector is a Postgres extension adding vector types and similarity search with HNSW and IVFFlat indexes. It makes vector search available inside an existing relational database rather than in a separate system.

## Where pgvector fits on a long engagement.

The architectural argument is strong for most products. Vectors beside the relational data means one backup, one set of transactional guarantees, and joins between embeddings and business data in a single query — no synchronisation, no consistency window, no second thing to operate.

Index choice is the substantive decision. HNSW gives better query performance and recall at the cost of build time and memory; IVFFlat builds faster and uses less memory with lower recall. Both are approximate, and the accuracy-versus-speed setting is a deliberate choice rather than a default to accept.

## What an assigned team does with pgvector.

The practical ceiling is higher than most teams expect. Millions of vectors with appropriate indexing perform well, and a great many products that adopted a dedicated vector database never needed one.

Establishing where that line sits for a specific workload is measurement rather than assumption, and it is part of the architectural work assigned under [ai engineering services](https://azendo.co/services/ai-engineering/).

## What we use pgvector for.

* Vectors and business data in one query Similarity joined to relational filters without crossing a system boundary.
* One thing to back up and operate No synchronisation path and no second consistency model.
* Index chosen deliberately HNSW or IVFFlat selected against measured recall and latency requirements.

## How pgvector capacity is assigned.

Vector search in Postgres is assigned across AI and data capacity on one agreement, since the decision spans both.

## Roles we assign pgvector for

* [AI Engineer AI engineering](https://azendo.co/services/ai-engineering/ai-engineer/)

## Service lines it sits in

* [AI engineering](https://azendo.co/services/ai-engineering/)

Capacity is agreed as a committed monthly capacity across a discipline, not per skill.

## Related in ai and machine learning

* [PyTorch — skill we assign for](https://azendo.co/skills/pytorch/)
* [TensorFlow — skill we assign for](https://azendo.co/skills/tensorflow/)
* [scikit-learn — skill we assign for](https://azendo.co/skills/scikit-learn/)
* [Pandas — skill we assign for](https://azendo.co/skills/pandas/)
* [NumPy — skill we assign for](https://azendo.co/skills/numpy/)
* [embeddings — skill we assign for](https://azendo.co/skills/embeddings/)
* [function calling — skill we assign for](https://azendo.co/skills/function-calling/)
* [vLLM — skill we assign for](https://azendo.co/skills/vllm/)

## Tell us what your roadmap needs pgvector for.

A service delivery manager replies with the disciplines we would assign, the monthly capacity and what the first month looks like.

[All skills we assign for](https://azendo.co/skills/)