Sofiia Alieva

Fabric Data Agents can query your semantic models, but can they truly understand your business? Semantic models describe the shape of your data, but they don't capture meaning. That missing layer is an ontology. This session exposes where agents break down and introduces practical patterns for enriching your semantic layer with ontological metadata: business glossaries, concept hierarchies, relationship semantics, and inference rules. Through live demos, you'll see how these enrichments help Data Agents resolve ambiguity, reason across domains, and produce trustworthy answers. You'll leave with a maturity framework and a concrete roadmap to get from "semantic model" to "knowledge layer" without boiling the ocean.

AI is everywhere—but what does it actually mean for your day-to-day work as a data engineer? In this session, we’ll cut through the hype and focus on practical, actionable ways to use AI tools in your workflow. Whether it’s generating and debugging Python code, or writing documentation, you’ll see how AI can help with real tasks—not just theoretical ones. We’ll also discuss when not to trust AI, how to avoid hallucinated outputs, and best practices for using these tools responsibly in production-focused environments.