Join us for an exciting demonstration on how to transform raw data in a database into a searchable format using Natural Language Processing (NLP). You will learn how to leverage Retrieval-Augmented Generation (RAG), vectors, NL2SQL, and agents, all within T-SQL, to create a powerful and scalable solution. Don't miss this opportunity to see how you can build a full-stack end-to-end solution in just 15 minutes!

To learn more, please check out these resources:
 * https://aka.ms/build25/plan/ai_agents_apps__data


𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀:
 * Davide Mauri


𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻:
This is one of many sessions from the Microsoft Build 2025 event. View even more sessions on-demand and learn about Microsoft Build at https://build.microsoft.com

DEM563 | English (US) | Data & Analytics


#MSBuild

Chapters:
0:00 - Introduction and Session Overview
00:00:07 - Demonstration on Building GNTKI Search in Azure SQL
00:04:22 - Decision Making in AI Query Execution
00:05:20 - Alternative Methods for Implementing AI Solutions in Databases
00:07:51 - Options for Using Different Technologies and Platforms
00:10:10 - Evaluation of Database Response and Query Accuracy
00:11:03 - Introduction to Advanced Code Sample Query of 2025
00:14:38 - Setup and initialization of configuration file
00:15:00 - Connecting to Azure SQL and Swagger setup

Microsoft Developer

The session explores how to build an agentic RAG solution using Azure SQL, OpenAI, and Data API Builder. It explains the process of embedding and semantic searches within Azure SQL and SQL Server 2025 for improved AI interactions with databases. By utilizing tools like TSQL and Data Builder, complex SQL queries can be orchestrated and executed efficiently, allowing SQL engines to securely manage AI-based solutions. The presentation also demonstrates how to turn stored procedures into REST or GraphQL endpoints seamlessly.

Build Agentic RAG solutions with Azure SQL OpenAI & Data API builder | DEM563