M365 Con Copilot & copilot Studio

Описание: 9 AM - Femke Cornelissen (MVP & Chief Copilot, Wortell)
Keynote: The Agentic Decade. How we redefine work, skills & leadership
The Agentic Decade is here and it’s transforming everything we know about work. We are moving from task-based productivity to a world where AI agents run processes, make decisions, and reshape how organisations operate. In this 30-minute session, Femke shares where the Agentic Workplace is heading, which skills will define the next generation of professionals, and why community and continuous learning matter more than ever. You’ll discover how roles evolve when AI becomes a teammate, how to design agent-ready workflows, and what leaders must do to stay ahead. A fast, practical and inspiring session for anyone who wants to stay relevant in a world where agents don’t just support your work.. they run it.

10 AM - Kevin McDonnell (Copilot Strategy and Modern Workplace AI Lead, Avanade)
Anatomy of an agent
What actually is an agent? While the term has been increasingly used across the industry, the reality of what it means can be very different, depending on who and when you speak to someone (especially Microsoft). The excitement of what can happen is often swiftly followed by a confusion of what will actually happen. This session will break apart the anatomy of what an agent is and talk through the different components of the modern, AI powered capabilities and show you what you need to consider. It will explain the building blocks needed and lead you towards how the decisions on those blocks can help you understand which technologies to use when. With each of these, you will see what different technologies are available and how you can choose what to use when. You will learn: - What are the building blocks of an agent - How do you design your agent to provide the right value - What technology is available to build agents for end users, power users and specialists - When you don’t need an agent - How to be a responsible agent builder I will talk focus on Copilot Studio agents, Copilot Studio Lite/Declarative Agents and Microsoft Foundry with a little of the SharePoint agents and may even sprinkle in a little Salesforce Agentforce. Remember, if your agent is not from the region of generative AI, it is just sparkling automation.

11 AM - Ragnar Heil (Microsoft MVP, HanseVision)
Copilot Agents with MCP Servers: The Good, The Bad and the Ugly
This session dives into the evolving landscape of Copilot Studio agents integrated with Model Context Protocol (MCP) servers. It highlights the advantages of connecting agents to external systems for richer, more contextual outputs that transform enterprise workflows. Participants will uncover common pitfalls in security, scalability, and governance when deploying these advanced agents in production. Real-life implementation stories will reveal both the brilliance and the breaking points of this new architecture. Attendees will leave with practical frameworks for balancing innovation, compliance, and control in MCP-powered Copilot ecosystems.

12 AM - Wario Wario (Senior Cloud Solution Architect Copilot, Sulava MEA)
Building a Robust Copilot Studio Strategy
As Copilot Studio adoption accelerates, organizations must move beyond ad-hoc agent creation and establish a well-architected, enterprise-grade strategy. This session dives deep into the technical foundations required to design, deploy, and operate Copilot Studio at scale. We will examine core architectural decisions, including environment and tenant strategy, Dataverse and data source design, authentication and authorization models, connector usage and isolation, and secure integration patterns with Power Platform, Microsoft 365, and external APIs. The session also covers agent lifecycle management, versioning, ALM pipelines, monitoring, telemetry, and compliance considerations in regulated environments. Using concrete architecture patterns and real-world scenarios, this talk provides hands-on guidance for building Copilot Studio agents that are secure by default, observable, governable, and resilient. Attendees will gain a practical reference architecture and decision framework to support production-grade AI agents aligned with enterprise security, governance, and platform best practices.