If your Copilot Studio solutions are still built around topics and Power Automate flows, chances are you’re designing chatbots, not agents. And the difference is far more than a name change. Agents don’t just follow scripts and branching workflows - they reason, adapt, and make decisions. Building agents requires a new way of thinking, new skills, and a clear understanding of how to apply generative AI to solve real business problems.

This session is your fast-track to making that shift. Drawing on real-world experience and lessons learned from implementing Copilot Studio agents, I’ll share what works, what doesn’t, and the patterns that set successful solutions apart.

You’ll discover what the “agent mindset” really means and why it matters for solution design. You’ll learn the core principles of knowledge management for AI, how to use knowledge effectively, and when topics still play a role. We’ll explore the more advanced AI capabilities like generative orchestration, and AI-driven tools such as prompts, MCP servers, and Computer Use Agents, that enable agents to go far beyond “if-this-then-that” logic.

Whether you're a maker or a decision maker, you’ll leave with a clear understanding of how to identify high-value agent use cases, practical design patterns and the mindset and skillset to start building smarter solutions in Copilot Studio.