PGVector: RAG in Postgres

Learn how to store, update, and delete embeddings in real-time using PostgreSQL with the PGVector extension, demonstrating user-specific RAG with FastAPI, LangChain, and OpenAI.

Overview

When building RAGs, a common pain point is: where do you store your embeddings? How do you update or delete them in real-time? What if you want isolated knowledge per user?
That’s where PGVector comes in — an open-source vector similarity search extension for Postgres that makes all of this way easier.
With PGVector, you can keep everything in your DB, with minimal infra overhead, real-time updates, and native Postgres support.
I’ll show how I used PGVector in a RAG system with user-specific context and real-time updates — all running with FastAPI, LangChain, and OpenAI.

Tech stack

Related projects

Automatic Embeddings Using Supabase

Dublin

Learn how to automatically generate and store vector embeddings in Supabase using Postgres triggers and pgvector whenever a…

Vector + Graph Friends

San Francisco

Shows a hybrid vector‑graph RAG system that creates personalized event emails using a knowledge graph and vector search,…

Omni ingestion RAG

Medellín

This talk covers multimodal ingestion in Retrieval Augmented Generation applications, focusing on processing unstructured data including images, tables,…

Enhancing AI with RAG - Techniques to improve accuracy

Bogotá

Shows how GraphRAG and context-aware RAG use knowledge graphs and contextual embeddings with graph databases and ranking techniques…

RAGBuilder by Krux AI

Bengaluru

Learn how to tune chunking strategies, sizes, and other RAG parameters, evaluate configurations on a test set, and…

Securing vector embeddings for Gen AI RAG applications with VectorX Db

Bengaluru

Learn how vector inversion can expose data, why query logs leak information, and how VectorX DB enables encrypted…