Track

AI

Artificial intelligence is no longer a future conversation, it is happening in every project, every team, and every tool. This track cuts through the noise with practical, honest sessions on Copilot, AI agents, and Copilot Studio. Come ready to learn, challenge assumptions, and leave with clarity on what actually works.

🧠 MCP or not to MCP - that is THE question

English · Advanced (300)

Daniel Laskewitz Daniel Laskewitz

So, the Model Context Protocol (MCP) is everywhere these days, and now that Copilot Studio jumped on board in March 2025, everyone's wondering: what's the big deal? And more importantly, what happens to all those connectors we've been using? Daniel's going to break it all down for you in this session. We'll dive into what MCP servers actually are, how you can start using them in Copilot Studio, and why you might want to try them out in other AI tools too. Fair warning: this is going to be demo-heavy, so get ready to see MCP in action across a bunch of different scenarios. What you'll walk away with: ✅ A solid grasp of what MCP servers are and why they're worth your time ✅ Practical know-how for getting MCP servers up and running in Copilot Studio ✅ Real examples of using both Microsoft's own MCP servers (like the Dataverse MCP server) and third-party ones in different AI tools Come ready to see lots of demos and leave with a clear understanding of whether MCP is right for your projects.

Agent Builder vs Copilot Studio: Choosing the Right Tool for AI Agents

English · Intermediate (200)

Andrew Bibby Andrew Bibby
Sebastian Sieber Sebastian Sieber

Microsoft has a long tradition of renaming products, and AI agents are no exception. What was briefly known as Copilot Studio Lite is now Agent Builder in Microsoft 365 Copilot, sitting alongside the full Copilot Studio experience. But what are the real differences, and how do you decide which one to use? In this session, Sebastian Sieber and Andrew Bibby will take you through a practical, hands-on comparison of Agent Builder and Copilot Studio. Using real-world examples and live demos, we’ll show what each tool is designed for, where their strengths lie, and where the limitations start to appear when requirements grow. Get ready for a jam-packed session filled with demos, expert tips, best practices for writing effective prompts, and guidance on deploying agents across different channels. We will also touch on the developer options available for building agents that go beyond Copilot Studio using tools like Microsoft Foundry. Whether you are new to AI agents or looking to enhance your skills, you'll clearly understand when to use Agent Builder, when to advance to Copilot Studio, and how to create AI agents that add value.

Balancing Compliance and Usability for AI Apps in M365

English · Intermediate (200)

Julian Kusenberg Julian Kusenberg
Luise Freese Luise Freese

AI features are rapidly being integrated into business apps built on Microsoft 365, Power Platform, and Azure AI Foundry. But as soon as these apps move beyond proof-of-concept, familiar questions arise:​ Who can see what? Where does the data go? How do we control the prompts? What does ‘compliant’ even mean in a Copilot interface?​ This session offers a practical guide to designing AI-powered business apps that respect enterprise compliance boundaries without sacrificing usability. You’ll learn how to classify data early, protect sensitive outputs, and avoid common pitfalls like overblocking or shadow AI.​ We’ll walk through real examples, such as prompt-based approvals, document classification helpers, or AI-powered knowledge apps, and show how to implement the right mix of Purview, role-based access, and in-app transparency. ​ The goal: business apps that are secure, traceable, and usable from day one.

Closing Note - Your agent is brilliant, and completely unhinged. Let’s fix that

English · Intermediate (200)

Scott Durow Scott Durow

Last year Scott introduced a lot of us to new hip young people terms like "Italian Brainrot", but he did also show the state of Vibecoding in 2025. Which was interesting but overall he seemed a bit underwhelmed. Well one year has gone and it feels like the world of AI has turn - a lot. So we asked Scott to come back and let's find out how his view on AI / Vibe coding has changed in a year!

From Clippy to Copilot - A Document Processing Journey

English · Intermediate (200)

Alex Dan Alex Dan
Denisa Mihai Denisa Mihai

Long-long time ago, in the early days of Word, a brave little paperclip asked, “It looks like you’re writing a document. Need help?” Well... we have come a long way from that moment, haven't we? Now we have AI that guides and supports our every step. In a world of Copilots, Document Processing is maybe the use case that benefits the most from the GenAI revolution. Best part? Low-code is the framework that makes it all achievable: big benefits with low effort. And things have gotten a lot easier lately! This session is a guided tour through Microsoft’s low code document processing universe. We’ll explore the tools in our toolbox: from Power Apps, Power Automate, and AI Builder, to the latest features in Copilot Studio. Expect: - Live demos, - Live building, - Questionable nerdy references 😁 This can run as a session or a hands-on workshop. If you're a dev, admin, solution architect, or just someone who’s had to wrestle with a .docx or pdf, this session is your destiny. Also, if you think Clippy deserved better... same. 🙂

How and why did we implement AI Chat in FetchXML Builder?

English · Intermediate (200)

Andreas Adner Andreas Adner
Jonas Rapp Jonas Rapp

There was a great, fun, learning, and productive collaboration with Jonas and Andreas when we started sharing and exchanging knowledge, and ultimately delivered a brand new feature in XrmToolBox's tool: FetchXML Builder - AI Chat. The primary purpose is that since AI is fused with us everywhere, we should embrace it, get to know it, own it, and use it. What did we build? What is the focus? What was struggling us? How can it improve our day-to-day? Why do I now even share my own Azure AI Foundry for free? I will provide you with an overall perspective, delving into the code behind the scenes, and demonstrating how we can utilize it in practice.

Microsoft Foundry: Make AI Work for Your Business

English · Intermediate (200)

Elvira Cekrezi Elvira Cekrezi
Ema Myrtollari Ema Myrtollari

AI isn’t just for tech experts, it can make your workday easier, faster, and smarter. In this session, we’ll show you how Microsoft Foundry can solve real business problems and help your team focus on what really matters. You’ll see how to: Turn data into decisions without spending hours analyzing spreadsheets Automate repetitive tasks like ticket sorting, approvals, or document processing Connect AI with your existing tools like Power Automate and Dynamics 365 Use insights to act fast, reduce mistakes, and save time This session is about practical examples, real projects, and tricks you can use immediately to make your business work better. Who should attend: Business users and leaders who want smarter, faster workflows Developers and analysts looking for hands-on AI tips Teams who want to save time, reduce errors, and get more done Format: Demos of real AI workflows in action Step-by-step walkthroughs of practical business scenarios

Mission: Extensibility - Copilot Agents & Connectors in Action

German · Intermediate (200)

Fabio Bonolo Fabio Bonolo

Copilot can do more than you think - if you know how! 💡 🚀 In this session, you'll get the briefing for your “Mission: Extensibility”. You will learn how to extend Microsoft 365 Copilot using agents, connectors and automations - without additional licensing costs. 💰 🔍 What you can expect: đŸ‘‰đŸ» Ways & possibilities to extend M365 Copilot đŸ‘‰đŸ» Insights into Agents, Connectors & Co. đŸ‘‰đŸ» Agent Builder, Copilot Studio & Admin Center đŸ‘‰đŸ» Live demo and use cases Whether you already have experience or are curious to get started - this session will provide you with impulses, inspiration and immediately applicable know-how. 🎯 đŸ§‘đŸ»â€đŸ’Œ Ready to turn your Copilot into a super agent? Then: mission accepted!

Rethinking Test Automation: Streamlining Agent Development with the Copilot Studio Kit (lowcode)

English · Intermediate (200)

Holger Imbery Holger Imbery
Jennifer Krause Jennifer Krause

Discover how the Copilot Studio Kit transforms test automation for custom agents built in Microsoft Copilot Studio. In this 45-minute session, you'll get a hands-on look at how this innovative toolkit simplifies and accelerates the development and testing process. What you’ll learn: - Introduction to the Copilot Studio Kit: A quick overview of its purpose and core capabilities. - Key Features: Deep dive into the unique functionalities that make the kit stand out. - Workflow Optimization: Live demonstrations showing how the kit enhances efficiency in development and testing. - Real-World Use Cases: Practical examples that highlight the kit’s impact across different scenarios. By the end of the session, you’ll understand how to leverage the Copilot Studio Kit to improve reliability, reduce development time, and scale your (lowcode) custom agent solutions with confidence. Target Audience: Developers, QA engineers, and technical leads working with Copilot Studio or exploring automated agent development. Level: Intermediate

These are (not) the agents you are looking for – Copilot Studio Projekte, die wirklich funktionieren

German · Intermediate (200)

Robin RosengrĂŒn Robin RosengrĂŒn

Ideen fĂŒr Agents gibt es in jedem Unternehmen wie Sand am Meer. Aber die wenigsten davon sind tatsĂ€chlich geeignet fĂŒr Copilot Studio. Gemeinsam schauen wir, was Copilot Studio kann und was nicht. Daraus leiten wir ab, welche Use Cases ihr euch als nĂ€chstes anschauen solltet und welche ihr erstmal auf die lange Bank schiebt. Am Ende erhaltet ihr klare Aussagen, welche Arten von Agents in Copilot Studio Business Value liefern und zudem noch einfach umgesetzt werden können.

Unscripted AI expert panel

English · Introduction (100)

Areti Iles Areti Iles
Dan Barber Dan Barber
Nathan Sweeney Nathan Sweeney
Scott Durow Scott Durow

Join four experts from both business and technical backgrounds for an open Q&A on AI. This is your chance to ask the questions that matter most to you. Expect honest, practical insights based on real experience, not just theory. Scott Durow - Microsoft Areti Iles - Telefonica Tech Dan Barber - Founder Strathos Nathan Sweeney - KPMG

Who's reading your prompts, and should you even care? Let's chat!

English · Intermediate (200)

Yannick Reekmans Yannick Reekmans

New AI tools like Claude Code, Copilot Cowork, and a growing ecosystem of LLM-powered copilots, along with an overload of AI-powered tools, promise to make us faster, smarter, and more productive overnight. And they often do. But behind the magic sits a question most of us don’t fully stop to ask: where does your data actually go? In this session, we dive into what really happens when you start using modern AI tooling that connects to multiple large language models — often with limited transparency around data handling, storage, and access. From prompts and uploaded files to context pulled from your tools, we’ll explore what you might be sharing (knowingly or not), and who could potentially see it. The question isn’t new, but are we now more willing to overlook the consequences because this technology is shiny, new, and incredibly powerful? Are we turning a blind eye to data privacy and governance, because if we don’t use them, we’ll fall behind? Let's explore together in this interactive, thought-provoking conversation, if our fears are irrational or maybe not big enough. Do we need to worry about GDPR or data residency with AI, and when exactly (not)? We might not get an answer to any of these questions, but you'll for sure go home with material to think about!

Will I really be needing a developer in the future?

English · Intermediate (200)

Sara Lagerquist Sara Lagerquist

Is this AI thing just low-code revolution 2.0, or are we looking at something that will actually change more than what we saw all those years ago? Because we’ve been here before. Promises of building faster, easier, with less reliance on developers. And yet, somehow, the need for developers didn’t go away. So what’s different this time? What happens when the line between “developer” and “builder” starts to blur even further? And are we actually shifting roles
 or just changing the tools? The day before this, I’ll be in a hands-on workshop, learning how to use AI to build things I normally hand over to developers. If that works
 things get interesting. Will I really be needing a developer in the future?