The mechanics of integrating AI and vector search into Azure Database for PostgreSQL
summary
Postgres on Azure stores 1,536‑dimensional embeddings with the PGvector type, generates them via the Azure AI extension calling ada-002, and indexes them with HNSW for fast nearest‑neighbor queries; an end‑to‑end demo app searches 120k listings, filters within 30 miles via PostGIS, vectorizes images, summarizes reviews with cognitive sentiment and abstractive summarization, and calls a Davinci generation endpoint to produce chat responses — all running in managed Azure Database for PostgreSQL with Entra ID, private links and customer‑managed keys.
AI-written summary · about
chapters (7)
- 0:00Postgres and AI overview
- 1:09Azure Postgres extensions
- 2:36Enterprise security features
- 3:39Building an AI search app
- 5:25Advanced SQL and indexing
- 7:54Sentiment and summarization
- 9:32Generative AI chat integration
related talks
27m6 things you can do with azure_ai & PostgreSQL on Azure
44mPostgres and OpenAI - a match made in Azure
47mExtendable by Design: Building Generative AI Apps with Postgres and Vector Storage and Azure AI
20mBuilding Enterprise RAG with Azure Database for PostgreSQL and pgvector
40mUnleashing the Power of Vector Search with Azure SQL Database
15mAzure Cosmos DB as Vector Database
01h 16mAzure SQL ❤️ Open AI
26mProduction RAG at Scale with Azure Database for PostgreSQL