Azure SQL ❤️ Open AI
summary
Live demo shows a full stack solution: Azure SQL issues REST calls to OpenAI to generate embeddings, stores them as varbinary (preview vector type coming), and performs similarity search; Azure Functions use SQL trigger binding to auto-generate embeddings on row changes; Data API Builder exposes stored procedures as REST/GraphQL for a React frontend; the pipeline uses vector search then GPT re-ranking (RAG) to keep costs down and returns human answers. Repo has main and classic branches and measured 29 ms average vector search at a past event.
AI-written summary · about
chapters (10)
- 0:00Azure SQL and OpenAI intro
- 5:59AI apps with your data
- 9:55Embeddings and vector search
- 21:10AI solution architecture
- 27:48Why vectors in SQL
- 31:20GitHub repo and setup
- 45:05RAG with prompt engineering
- 55:43REST from stored procedures
- 1:03:28Q&A
- 1:12:02Closing and developer feedback survey
related talks
46mPower AI apps and develop rich experiences with Azure SQL Database
50mBuilding AI ready applications
40mUnleashing the Power of Vector Search with Azure SQL Database
44mPostgres and OpenAI - a match made in Azure
27m6 things you can do with azure_ai & PostgreSQL on Azure
50mHow to design and build AI applications with vector search using Azure OpenAI & Azure Cosmos DB
12mBuilding Generative AI Apps with Azure Cosmos DB
12mThe mechanics of integrating AI and vector search into Azure Database for PostgreSQL