Momentum One Academy · lab

Chat with Your Data Using Microsoft Fabric Data Agents

This intermediate lab explores how a Fabric data agent can make a sales Warehouse accessible through business-language questions. Participants connect the agent to a prepared dimensional dataset, ask analytical questions, and review the SQL generated behind the response.

What the lab covers

Connecting a Fabric data agent to a retail star-schema Warehouse.
Asking questions about sales, product and customer information.
Reviewing generated SQL to understand selected tables, joins and aggregations.
Interpreting the results in the context of business definitions, fiscal-year conventions and currency.

Learning outcome

Participants can explain how a data agent uses connected structured data to answer natural-language questions and why generated SQL should be checked. They also learn to confirm dataset-specific calendar and currency semantics before presenting an AI-generated result as a business fact.

Workshop connection

This overview supports Module 08 — Direct Lake, Semantic Models, Copilot & Data Agents. It follows the semantic-model readiness discussion with an agent-based consumption pattern and a practical validation mindset.

Access note

The lab requires a paid Fabric capacity with the necessary Copilot/data-agent features; trial capacity is not supported. The source guide notes that generated SQL and sample answers can vary by environment. The training supplies the setup materials and detailed instructions; none are published on this page.

Further learning

Plan your workshop