NK.
All work

Case study 05 of 08Complete2025

FabCon25 Europe 2025 Talk to Your Data Demo

Talk to Your Data. A Fabric demo with Copilot and Data Agents.

Demo integrating Lakehouse data, semantic models, Power BI Chat with Data, in-report Copilot, Data Agent RCA, Copilot Studio orchestration, and JIRA MCP incident context.

Role
Demo architect
Conversational BI, governed context, root-cause flow
Stack
Fabric + Copilot
Power BI Chat with Data, Data Agent RCA, Copilot Studio
Scale
Conference demo
Public-safe talk-to-your-data concept
Outcome
Executive-friendly analysis
Natural-language path from question to BI-backed answer
01

The brief. What it took, and what it changed.

Problem

Show how business users can move from data questions to root-cause and incident context through Microsoft Fabric, Power BI, and agentic workflows.

System

Fabric Lakehouse and semantic model connected to Power BI conversational analytics, Copilot Studio orchestration, data-agent RCA, and JIRA MCP incident context.

Impact

Team solution recognized by senior company and Microsoft leadership for applied Fabric, Power BI, Copilot, and data-agent capabilities.

02

How it works. In 3 stages.

    01

    Lakehouse context

    Fabric data and semantic models give the conversational BI experience a governed analytical base.

    02

    Conversational BI

    Power BI Chat with Data and in-report Copilot create the business-user question path.

    03

    Root-cause layer

    Data Agent RCA and Copilot Studio connect analysis to incident context and next actions.

03

What shaped the build. And what it runs on.

    Position as a demo recognition moment rather than a standalone product until more public media is available.

Built with

  • Microsoft Fabric
  • Power BI
  • Copilot
  • Copilot Studio
  • Data Agent
  • JIRA MCP
04

Proof. What's public, and what's on its way.

References

  • Demo referencePublic event or demo reference to add after approval.Pending

Visual artifacts

  • Demo screenshotsscreenshotTo approve
  • Architecture mapdiagramTo approve
  • Event referencedocumentTo approve
05

Tradeoffs. And what stays private.

The tradeoff

Strong support for AI-assisted analytics, but public assets should be approved before it carries major homepage weight.

Kept anonymous

Confirm demo assets and company/client references before public use.

Next case studyInsurance analytics. Semantic models, with an AI assistant on top.Read it