What happens when you turn the tables on your own AI? In this session, we dive into the fascinating journey of building an autonomous agent with Copilot Studio—and then trying to break it. You’ll hear the real story of how an agent designed to streamline job applications was stress-tested with prompt injections, tool misuse, and data exfiltration attempts. Discover where the agent held strong, where it failed spectacularly, and what lessons emerged for anyone building AI-driven automation. This is not theory - it’s a candid look at the vulnerabilities, defenses, and design patterns that matter when your agent moves from the lab to the real world. Why Attend? If you’re working with Copilot Studio, Power Automate, or planning to deploy autonomous agents, this session will give you practical insights into: • Common attack surfaces and how to mitigate them • Designing safe, auditable actions and flows • Grounding AI responses to enterprise data • Building resilience without sacrificing usability Join us for a story that blends innovation, security, and a little bit of hacking fun—because the best way to understand your agent is to try breaking it.

Responding to RFPs is a complex, high pressure process that rarely fits into a single workflow—or a single AI agent. Requirements analysis, compliance checks, content drafting, and consistency reviews all demand different skills and perspectives. In this session, I present Proposal Operator, a practical case study of using multiple Copilot Studio agents working together to support RFP responses. Instead of building one “do everything” agent, the solution is based on specialized agents with clear responsibilities, orchestrated as a team to handle different parts of the process. The focus is not on clever prompts, but on architecture and orchestration: when one agent delegates work to others, how responsibilities are separated, and how the system stays understandable and governable as complexity grows. The RFP scenario serves as an example of a broader pattern relevant to many enterprise processes where work must be split, reviewed, and coordinated. You’ll see why multi agent designs are often more robust than single large agents, what breaks when orchestration is missing, and how Copilot Studio can be used to model agent teamwork in a controlled way. This session is for architects and developers who want to move beyond single agent demos and start designing agent systems that scale in both capability and complexity.

In automation, actions aren't just louder—they're smarter. This session walks through Power Automate patterns drawn from a real, fully-working demo: a Copilot Studio job-application assistant and the five Power Automate flows that run everything after someone applies. Karl-Johan Spiik will show you how disciplined patterns—iterative development, proper error handling, smart child-flow design—turn a fragile flow into one you can actually trust in production. And in the session's closing stretch, he'll show what happened when he handed that same solution to an AI coding agent and asked it to find what he'd missed. What You'll Learn - Build flows iteratively, one action at a time, so failures are easy to isolate and debug - Apply a proper try/catch scope pattern, and know when to terminate a flow with a Failed status so it doesn't silently vanish into "Succeeded" - Move logging from ad-hoc checks to a real production tier with Application Insights via the Power Platform Admin Center - Replace hard-coded approval lists with a Teams-channel-membership pattern that scales without spamming anyone - Scope child-flow permissions with run-only users, and respond early from slow child flows so the parent isn't left blocked - Build a self-service Power Platform Lifecycle Management portal — automate environment creation, connector unblocking, and Entra ID security group provisioning end to end, with SharePoint (not Dataverse) as the intake form for licensing reasons - Use an AI Prompt action directly inside Power Automate — no full Copilot Studio orchestration needed - See a live case study: an AI coding agent (Claude Code) given the same demo and existing conventions, autonomously finding and building five missing flows via the Power Platform CLI Why Attend? Karl-Johan Spiik brings nearly two decades of Microsoft project delivery experience, including eight years focused on Power Platform, and has spent the past few months building agent-driven automation with Claude Code. Join this session and learn how to make your flows Spiik louder than words — because in Power Automate, actions truly speak louder.

Copilot Studio agents can produce impressive answers—but impressive is not the same as trustworthy. In enterprise use, the real challenge is not getting good responses, but making sure those responses are predictable, sourced, up to date, and safe to rely on. In this session, I focus on designing Copilot Studio agents that stay within clear boundaries: what they are allowed to answer, where their knowledge comes from, and when they should not answer at all. Rather than treating AI as a black box, the session looks at governance as a design concern, built into the agent from the start. You’ll see practical patterns for controlling knowledge sources, avoiding stale or unintended information, and designing agents that behave consistently across changing data and use cases. The emphasis is on builder level decisions—how structure, scoping, and governance choices directly affect trust in AI generated answers. This session is not about making agents sound smarter, but about making them defensible in real business scenarios. This talk is for architects, developers, and platform owners who want to move from “smart answers” to AI solutions that users and organizations can actually trust.

Power Platform and Copilot agents enable rapid innovation—but without the right governance, they can introduce security risks, operational friction, and long-term technical debt. This session shows how governance can accelerate adoption instead of slowing it down. Using real-world enterprise patterns, we present an end-to-end governance model covering the full lifecycle from idea intake to production, operations, and ongoing support. Attendees will learn why documentation alone is not enough, and how operational governance helps balance business agility with enterprise IT control. The session introduces a practical governance add-on approach that complements existing ITSM tools, providing a single intake for environment requests, connector approvals, access management, and Power Platform or Copilot Studio agent issues. We also cover approval and integration models—including hybrid and tiered approvals—and explain why these matter in customer-facing discussions. Finally, we look at the Copilot Studio agent lifecycle: ownership, citizen vs. IT delivery paths, quality gates, controlled releases, and the often-overlooked challenges after go-live. Who should attend: Sales, solution architects, delivery leads, project managers, and Power Platform or Copilot Studio specialists. Key takeaway: A clear, practical way to position governance as a business enabler.