#multi-agent

Sep 3, 2026

A Data Agent Should Be a Sub-Agent

Here’s how most data agents get built. Someone connects one to a semantic model, opens the big chat pane in the middle of the screen, and types “how are sales?” A number comes back. Everyone nods. Nobody opens it again. That’s what most people think a data agent is for: another chat interface. One more box to type into, sitting next to Copilot, ChatGPT, Claude, and whatever your company bought last quarter. I think a data agent should be a sub-agent. It’s a narrow specialist with one job written into its instructions, and a bigger agent calls it when a question needs it. People talk to Copilot, inside the tools they already work in. Copilot calls the data agent. Once you look at it that way, two things change. You stop building one data agent per semantic model, and you start writing instructions per purpose.

Dec 20, 2025

Wiring AI Agents to Enterprise Data with Azure AI Foundry + Fabric

This demo shows one concrete way to design a multi-agent system using Azure AI Foundry and Microsoft Fabric.

Data Agent Client - Teams App

A Microsoft Teams application that enables natural language queries to Fabric Data Agents, featuring real-time streaming responses, DAX query visualization, and multi-agent support. - Natural Language Queries: Ask questions in plain language and get AI-powered answers from your semantic models - Multi-Agent Support: Connect to multiple Fabric Data Agents and switch between them seamlessly - Real-Time Streaming: See responses as they're generated with live progress indicators - Query Transparency: View the generated DAX code and query results for each analysis step - Teams Integration: Native Teams app experience with SSO authentication - Fluent UI Design: Modern, responsive interface following Microsoft design guidelines

Mar 16, 2026

Multi-agent routing for Fabric Data Agents using Microsoft Foundry

Creating multi-agent architecture with data agent and Microsoft Foundry

Jul 20, 2026

The Fabric Data Agent Routing Challenge and the Architecture That Solved It

We recently encountered an important architectural constraint while integrating Microsoft Fabric Data Agent with Microsoft Copilot Studio. Our use case was straightforward: Business users needed to ask questions about supplier, procurement, invoice, and spend data. At the same time, they also needed to trigger operational processes such as supplier onboarding, purchase approvals, support requests, and data issue reporting. The post shows how to separate analytical vs operational flow in Copilot studio.