Building Enterprise RAG with Azure Database for PostgreSQL and pgvector
summary
Shows a working RAG stack on Azure Database for PostgreSQL using the pgvector extension, the create_embeddings wrapper tied to Azure OpenAI, and SQL functions that batch-create and store embeddings. Demonstrates three index types—IVFFlat HNSW DiskANN—with a 76× speed gap observed between HNSW and IVFFlat on a tens-of-thousands corpus, and DiskANN outperforming HNSW when category filters are applied. All demos run end-to-end and include performance comparison queries.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 0:32Enterprise Challenges
- 2:20Native Vector Search
- 5:00DEMO - Vector Search
- 5:10Installing Prerequisites
- 8:09DEMO - Embeddings
- 11:20DEMO - Indexing
- 13:21Indexing Comparision
- 15:44DiskANN Demo
- 17:20Semantic Search
related talks
26mProduction RAG at Scale with Azure Database for PostgreSQL
12mThe mechanics of integrating AI and vector search into Azure Database for PostgreSQL
47mExtendable by Design: Building Generative AI Apps with Postgres and Vector Storage and Azure AI
44mPostgres and OpenAI - a match made in Azure
27m6 things you can do with azure_ai & PostgreSQL on Azure
46mTransform applications with AI and Azure Database for PostgreSQL
01h 16mAzure SQL ❤️ Open AI
01h 35mTutorial: Build AI-Supercharged RAG Apps with a Vector Database