Prepare AI-ready analytics data in Microsoft Fabric

At a glance

Prepare your semantic layer for AI by adding metadata and linguistic context to gold-layer data stores and semantic models, then generate ontologies that AI agents use to answer business questions.

Prerequisites

  • Experience designing and publishing semantic models in Microsoft Fabric
  • Familiarity with DAX measures and relationships in semantic models
  • Basic understanding of how AI agents consume structured data

Modules in this learning path

Design gold layers, semantic models, and documentation that enable Copilot, data agents, and enterprise ontologies to deliver accurate, business-relevant insights.

Microsoft Fabric IQ is part of Microsoft IQ, an enterprise intelligence layer that elevates business data into shared concepts usable by people and AI agents. Learn about the six items in the IQ workload—ontology items, data agents, Graph in Microsoft Fabric, Power BI semantic models, operations agents, and planning—and discover how ontology modeling starts with business concepts rather than specific use cases.

Ontologies in Fabric IQ transform your data into a business vocabulary that everyone can understand. This module covers two ways to create ontologies—building manually to understand the core components, or generating automatically from Power BI semantic models to accelerate development. You practice both approaches and learn how to connect your ontology to OneLake data sources, including lakehouse tables and eventhouse streams.