Build AI Apps with Oracle AI Database@Azure, MCP, and GitHub Copilot | DEMSP382
Parthasarathy Srinivasan and Rajya Laxmi Yellajosyula demonstrate a multi-cloud, enterprise AI workflow that combines Oracle Database@Azure with Microsoft Fabric, MCP, and GitHub Copilot, covering provisioning, synthetic data creation, ETL from bronze to gold, and an end-to-end fraud detection demo driven by natural-language orchestration.
Overview
This Microsoft Build 2026 demo session walks through building a secure enterprise AI workflow that spans Oracle and Microsoft platforms:
- Data platform: Oracle Database@Azure and Oracle AI Database
- Analytics/engineering: Microsoft Fabric
- Orchestration: Model Context Protocol (MCP), including identifying and using MCP servers/tools
- Developer assistance: GitHub Copilot in VS Code to automate and validate steps
The scenario centers on setting up a fraud detection example, including provisioning and data preparation, then using Copilot + MCP to reduce custom integration work when orchestrating tasks across Oracle and Fabric.
Session chapters (from the video description)
Session introduction and overview (0:00)
- High-level framing of the demo: enterprise AI workflow across Oracle and Microsoft Fabric using MCP and GitHub Copilot.
Demo 1: Fraud detection system setup (00:03:22)
- Introduces the first demo example and the overall fraud detection setup.
Creating synthetic data and automating SQL with Copilot (00:06:08)
- Creates synthetic data tables.
- Uses GitHub Copilot to help automate SQL-related work.
Demo 2: Autonomous Database integration (00:11:16)
- Introduces a second demo and reviews integration with an Autonomous Database.
Copilot identifies Fabric MCP server and tools (00:14:14)
- Uses GitHub Copilot to identify a Fabric MCP server and the relevant tools available through it.
Data engineering: ETL from bronze to gold (00:15:59)
- Switches to a data engineer role and builds an ETL pipeline moving data through bronze → gold layers (medallion-style layering).
Validating a REST API and setup in VS Code with Copilot (00:19:56)
- Validates a REST API.
- Uses VS Code with GitHub Copilot for initial setup and workflow assistance.
Building a semantic model and data agent in Fabric (00:21:02)
- Builds a semantic model.
- Creates a data agent for business users using Microsoft Fabric tools.
Microsoft and Oracle collaboration (00:24:01)
- Wrap-up discussion on the collaboration and how it supports data-driven decision making in enterprise scenarios.