Holger Imbery
Microsoft MVP Business Applications - Copilot Studio & MCT & PowerPlatform | Azure | Dynamics - principal architect @NTTData Inc.
Microsoft MVP Business Applications - Copilot Studio & MCT & PowerPlatform | Azure | Dynamics - principal architect @NTTData Inc.
Berlin, Germany
Actions
Holger is an Principal Architect at NTT DATA Inc., based in Berlin, Germany. He specializes in enterprise AI, Microsoft Copilot Studio, Power Platform, Dynamics 365, and cloud-based automation solutions.
His work focuses on designing large-scale business transformation projects that help organizations streamline operations, improve customer experiences, and adopt AI responsibly.
With extensive experience in enterprise architecture, agentic AI, and Microsoft technologies, Holger helps customers turn innovative concepts into production-ready solutions that create measurable business value. He regularly shares his insights through conferences, community events, and his blog at holgerimbery.blog
Holger ist Principal Architect bei NTT DATA Inc. mit Sitz in Berlin. Sein Schwerpunkt liegt auf Enterprise AI, Microsoft Copilot Studio, der Power Platform, Dynamics 365 sowie cloudbasierten Automatisierungslösungen.
In seiner täglichen Arbeit konzipiert und gestaltet er groß angelegte Transformationsprojekte, die Unternehmen dabei unterstützen, Prozesse effizienter zu gestalten, Kundenerlebnisse zu verbessern und Künstliche Intelligenz verantwortungsvoll einzusetzen.
Mit langjähriger Erfahrung in Enterprise-Architekturen, agentenbasierten KI-Lösungen und Microsoft-Technologien begleitet Holger Kunden dabei, innovative Ideen in produktionsreife Lösungen zu überführen, die einen messbaren geschäftlichen Mehrwert schaffen. Seine Erfahrungen und Erkenntnisse teilt er regelmäßig auf Konferenzen, bei Community-Veranstaltungen sowie auf seinem Blog unter holgerimbery.blog.
Links
Area of Expertise
Topics
Everybody Builds Copilot Studio Agents. Few Ship Them.
From Maker Success to Enterprise Success: ALM and GitHub Actions for Microsoft Copilot Studio Agents
Building a Copilot Studio agent has never been easier.
With Microsoft Copilot Studio, makers can create conversational experiences, automate business processes, integrate enterprise systems, and deliver business value in a matter of hours. As a result, many organizations are successfully building agents. The real challenge begins when those agents need to move beyond a single environment and become production-ready business solutions.
The challenge is no longer how to build a Copilot Studio agent.
The challenge is how to ship, govern, and operate Copilot Studio agents at enterprise scale.
In this practical, experience-based session, we'll explore what happens after the proof of concept. Drawing on real-world projects, lessons learned, and enterprise delivery patterns, we'll walk through the complete lifecycle of a Copilot Studio agent, from initial idea to production deployment.
Together, we'll follow the journey of a production-ready agent:
Identifying the right business scenario
Designing maintainable Copilot Studio agent architectures
Integrating business systems through connectors, APIs, and MCP
Managing agent assets using solutions and source control
Implementing Application Lifecycle Management (ALM) practices
Automating deployments with GitHub Actions
Managing environments, secrets, and configuration
Establishing governance, security, and compliance controls
Monitoring adoption, performance, and business value
Operating and continuously improving agents in production
Along the way, we'll discuss common mistakes organizations make when moving from demos to enterprise solutions, including manual deployments, poor environment strategies, missing source control, governance gaps, unclear ownership models, and treating agents as experiments rather than business-critical applications.
You'll learn how successful organizations apply the same engineering discipline used for modern software delivery to their Copilot Studio agents, using ALM, GitHub, and GitHub Actions to create repeatable and reliable deployment processes.
You won't leave with another demo.
You'll leave with a practical framework for delivering Copilot Studio agents that can be trusted, supported, and scaled across the enterprise.
One Goal, Two Paths - Choosing the Right Architecture for AI-Powered Voice Self-Service
Organizations want to offer conversational, AI-powered voice self-service, but choosing the right architecture is not straightforward. Do they need a full contact center platform, or can a more focused solution built with Teams Phone Agents and Copilot Studio meet the same business need?
In this demo-driven session, we will solve one customer service scenario in two different ways, both using ServiceNow as the backend platform:
- Dynamics 365 Contact Center with Copilot Studio voice agents
- Teams Phone Agents built with Copilot Studio
You will see how each approach handles the customer conversation, knowledge retrieval, ticket creation, escalation, and transfer to a human agent. We will then compare the architectures and discuss their strengths, trade-offs, complexity, and extensibility.
This is not a competition with a predetermined winner. By the end of the session, you will have a practical decision framework for evaluating both approaches and choosing the path that best matches your organization's requirements, existing platforms, and AI-powered voice strategy.
Multi-Agent Orchestration in Copilot Studio
Practical Patterns for Child Agents, Connected Agents, A2A, MCP, and Long-Running Business Processes
Multi-agent solutions are becoming one of the most powerful capabilities in Microsoft Copilot Studio.
But most teams quickly discover that creating multiple agents is easy. Designing an orchestration model that remains understandable, governable, and maintainable is much harder.
In this practical session, we'll explore how to build multi-agent solutions in Copilot Studio using the newest orchestration capabilities, including child agents, connected agents, Agent-to-Agent (A2A) communication, MCP integrations, and the emerging GitHub Copilot harness capabilities.
Rather than focusing on individual features, we'll focus on architecture and orchestration:
* When a single agent is enough
* When to introduce child agents
* When connected agents become the better choice
* How A2A changes cross-agent collaboration
* How MCP connects agents to tools and external capabilities
* How state-driven orchestration enables long-running business processes
* How to avoid creating agent architectures that become impossible to govern
We'll also examine the evolving Copilot Studio runtime landscape and explain where the Standard Harness and GitHub Copilot Harness fit into modern agent architectures. Rather than treating harness selection as the primary decision, we'll show how business requirements, orchestration patterns, and required building blocks naturally guide the choice.
Using real-world examples and implementation patterns, including long-running enterprise workflows, you'll learn how successful organizations move beyond simple copilots and build orchestrated agent ecosystems that solve complex business problems.
You won't leave with another collection of features.
You'll leave with practical orchestration patterns you can apply immediately in your own Copilot Studio solutions.
What You'll Learn
* Designing Multi-Agent Systems
* Identifying meaningful boundaries between agents
* Specialization versus reuse
* Avoiding unnecessary orchestration complexity
* Establishing ownership and governance boundaries
* Child Agents and Connected Agents
* When to use child agents inside a solution
* When connected agents provide a better architectural boundary
* Delegation patterns and specialist-agent design
* Managing shared and isolated responsibilities
MCP and A2A
MCP as the standard mechanism for exposing tools and resources
A2A as the standard mechanism for collaborating with autonomous agents
When to choose a capability call versus agent delegation
Combining MCP and A2A in the same architecture
Orchestration Patterns
Orchestrator + specialist agents
Specialist networks
Hierarchical orchestration
Event-driven orchestration
Long-running business processes
Harnesses in Context
A practical overview of:
Standard Harness
GitHub Copilot Harness
Child Agents
Skills
Memory
Native file handling
Connected Agents
with a focus on how these building blocks affect architecture decisions rather than feature comparisons.
Real-World Example
A detailed walkthrough of a state-driven invoice processing solution built with:
Copilot Studio
Child Agents
Agent Flows
Dataverse
Connectors
showing how orchestration, state management, and specialization work together in production architectures.
Key Takeaways
After attending this session, you will be able to:
* Design practical multi-agent architectures in Copilot Studio
* Choose between child agents, connected agents, MCP, and A2A
* Apply proven orchestration patterns to complex business scenarios
* Build long-running processes using state-driven orchestration
* Understand where the Standard Harness and GitHub Copilot Harness fit within multi-agent architectures
* Avoid the most common multi-agent design mistakes
* Create solutions that remain maintainable, governable, and scalable
Your Microsoft Copilot and Your Agents Need a Budget
Agent FinOps for Microsoft Copilot, Copilot Studio Agents, and Enterprise Agent Ecosystems
Microsoft Copilot, Copilot Studio agents, MCP servers, autonomous agents, connected agents, agent flows, and multi-agent systems are transforming how organizations work, automate processes, and create business value.
But as AI moves from experimentation to enterprise-wide adoption, a new challenge emerges:
How do we ensure that Microsoft Copilot and agent investments deliver sustainable business value?
Every design decision has consequences.
Deploying Microsoft Copilot for employees creates one economic model. Building Copilot Studio agents creates another. Introducing autonomous workflows, voice experiences, multi-agent orchestration, MCP integrations, connected agents, or Agent-to-Agent (A2A) collaboration introduces entirely new consumption, governance, and operational considerations.
In this session, we'll introduce Agent FinOps, a practical framework for evaluating, governing, and optimizing AI investments across the Microsoft Copilot ecosystem.
Rather than focusing solely on cost reduction, we'll explore how architects, platform owners, and business leaders can balance:
- Business value
- User adoption
- Governance
- Performance
- Scalability
- Consumption and licensing
Together, we'll examine the economics of:
- Microsoft Copilot
- Copilot Studio agents
- Autonomous and multi-agent ecosystems
and show how each creates fundamentally different operating, consumption, and value models.
Through practical examples and real-world implementation scenarios, we'll discuss:
Why every Copilot and agent architecture has an economic impact
How different agent architectures influence consumption and business value
The hidden costs of unnecessary complexity and over-engineered orchestration
Consumption models across Microsoft Copilot and Copilot Studio
Measuring value beyond simple usage and adoption metrics
Showback and chargeback approaches for AI investments
Governance as a FinOps capability
Operating large portfolios of copilots, agents, and autonomous systems responsibly
Because successful AI initiatives are not determined by the number of copilots, agents, or automations deployed.
They are determined by the value they create.
You will leave with a practical framework for making better architectural decisions, governing AI investments, and scaling Microsoft Copilot and agent ecosystems sustainably across the enterprise.
Key Takeaways
After attending this session, you will be able to:
- Understand the principles of Agent FinOps and how they extend traditional Cloud FinOps
- Differentiate the economic models of Microsoft Copilot, Copilot Studio agents, and autonomous agent ecosystems
- Identify architectural decisions that drive consumption, licensing, and operational costs
- Understand how MCP, agent flows, connected agents, A2A collaboration, and multi-agent orchestration influence economics at scale
- Measure business value beyond adoption and usage metrics
- Balance innovation, governance, security, and consumption within a unified operating model
- Design economically sustainable Copilot and Copilot Studio solutions
- Build showback and chargeback approaches that encourage innovation while maintaining accountability
- Create a roadmap for governing enterprise-scale portfolios of copilots, agents, and autonomous systems
Nordic Summit 2026 Sessionize Event
European Power Platform Conference (EPPC26) Sessionize Event
ColorCloud Hamburg 2026 Sessionize Event
Global Power Platform Bootcamp 2026 | Zürich Edition - #GPPBZH Sessionize Event
Dynamics Summit 09. & 10. Dezember 2025 Sessionize Event
Microsoft AI Tour
Unleashing Autonomous Agents with Copilot Studio: From Prompts to Purpose
Session code: LTG188-DE
We'll explore how Copilot Studio empowers creators to build intelligent, goal-driven autonomous agents that go beyond simple task automation. Covering what autonomous agents are and how they differ from traditional bots, key components in Copilot Studio for building agents, and a real-world use case. Whether you're a seasoned maker or simply curious about the future of AI-driven workflows, this session will inspire ideas and demonstrate how to start working with autonomous agents in Copilot Studio.
Power Summit 15. + 16.07.2025 Sessionize Event
Holger Imbery
Microsoft MVP Business Applications - Copilot Studio & MCT & PowerPlatform | Azure | Dynamics - principal architect @NTTData Inc.
Berlin, Germany
Links
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top