6 things you can do with azure_ai & PostgreSQL on Azure
summary
Shows concrete steps to add vector search and AI to Azure Database for PostgreSQL using the azure_ai and azure_local_ai extensions plus pgvector, with SQL UDFs that call Azure OpenAI deployments (text-embedding-3-small) or run a multilingual-e5-small ONNX model in‑database. Demo measures 100 embeddings at 1.3s for local vs 6.2s remote and compares TPS (local ~242 TPS, remote ~17 TPS, 4ms vs 61ms latency). Covers sentiment, summarization, real‑time translation (100+ languages) and azure_ml.invoke real-time scoring for a housing model; also shows create embeddings in SQL and in-database ONNX embeddings.
AI-written summary · about
chapters (12)
- 0:00Intro
- 4:56Create embeddings with Azure OpenAI
- 7:02In-database create embeddings
- 9:01DEMO: Generate embeddings within the database
- 13:58DEMO: Same scenario with local embeddings
- 16:37Sentiment analysis
- 17:23Summarization & phrase extraction
- 17:59Real-time text translation
- 19:31DEMO: AI language & translation
- 22:26Real-time predictions with Azure Machine Learning
- 23:19DEMO: Azure Machine Learning integration
- 26:50Resources
related talks
12mThe mechanics of integrating AI and vector search into Azure Database for PostgreSQL
44mPostgres and OpenAI - a match made in Azure
01h 16mAzure SQL ❤️ Open AI
48mThe power of AI and Copilot for Azure Databases
46mTransform applications with AI and Azure Database for PostgreSQL
40mUnleashing the Power of Vector Search with Azure SQL Database
20mBuilding Enterprise RAG with Azure Database for PostgreSQL and pgvector
15mAzure Cosmos DB as Vector Database