Developer & Speaker · Agentic Coding
I build & teach
Agentic Coding
I'm Dr. Matthias Liebeck, a senior solution architect, dual Microsoft MVP for Microsoft Foundry and Microsoft Azure, community speaker and avid GitHub Copilot power user. I work directly with agentic coding, the Model Context Protocol and AI agents, and I love sharing what I learn.
GitHub Copilot
Model Context Protocol
Agentic Coding
.NET & C#
Azure
Microsoft Agent Framework
// On stage
Talks
I speak at conferences, Microsoft communities and user groups across the German-speaking tech scene — mostly in German — about agentic coding, GitHub Copilot and MCP.
Upcoming talks
Catch me live
Featured talks
01
Bring Your Own Model: GitHub Copilot with Microsoft Foundry, OpenRouter and Local LLMs
Since June, every Copilot interaction is billed by tokens, and the model behind the session is the biggest cost lever. This talk covers the four ways to bring your own model into GitHub Copilot: directly from a provider, locally, via OpenRouter, and through a deployment in Microsoft Foundry. The focus is on Foundry: your own tenant, your own region, billing as regular Azure consumption, and why the serverless filter in the model catalog is the decision that matters most.
Azure Cloud Native - Köln · Cologne · 13.10.2026
📑 Slides (DE)
📑 Slides (EN)
🔗 Event
02
Build Your Own Agent with the GitHub Copilot SDK and Microsoft Foundry
The engine behind GitHub Copilot CLI is available as an SDK you can embed in your own .NET applications. In this session, I use it to build a real agent: a research radar that monitors the web for reports about queues and waiting times, for an actual product. Stage one is the core anyone can rebuild: the SDK loop, two small custom tools, file-based memory, and email notifications, running on your existing Copilot subscription. Stage two is the showcase: human feedback via vote links and an Azure Function, versioned agent memory in Azure Blob Storage, a scheduled container job on Azure Container Apps, and swapping the model underneath via Bring Your Own Key with a Microsoft Foundry deployment in my own Azure tenant. Real code, measured token costs, honest limitations, and a clear picture of when the SDK is enough and when you want a framework instead.
The Agentmakers · Virtual · 07.10.2026
📑 Slides
03
GitHub Copilot App: From Doing the Work to Directing It
Coding agents can do real work. The interesting question is what that does to your role: from writing every line yourself to deciding what gets delegated, what stays with you, and when to step in. In this talk, I show how the GitHub Copilot App is built for exactly this shift: parallel agent sessions in isolated git worktrees, autonomy as a dial from interactive to autopilot, automations for recurring work, and canvases where you and the agent work on shared state, built live from a single prompt. Plus a look at bring your own key, from OpenRouter to a model in your own Azure tenant. A hands-on introduction with honest notes on what is still rough, so you can judge what the app could take off your plate.
04
From Prompts to Skills: Advanced GitHub Copilot Workflows
Writing prompts is the easy part. The more interesting question is what comes next: How do you turn individual prompts into workflows that are reusable, remember past sessions, and critique their own work before any code is written? In this talk, I show the GitHub Copilot concepts I use every day: Agent Skills that capture project-specific knowledge like migration or test workflows once and make it available in every session. The Memory Bank pattern, which lets Copilot retain architectural decisions across sessions, transparently in Git instead of a cloud black box. And the rubber duck agent, where a model from a different provider reviews the plan before the first line is written. Plus an honest look at custom agents, Bring Your Own Model with local LLMs, and what the new token-based billing means for all of this.
05
From Prompts to Skills: Advanced GitHub Copilot Workflows
Writing prompts is the easy part. The more interesting question is what comes next: How do you turn individual prompts into workflows that are reusable, remember past sessions, and critique their own work before any code is written? In this talk, I show the GitHub Copilot concepts I use every day: Agent Skills that capture project-specific knowledge like migration or test workflows once and make it available in every session. The Memory Bank pattern, which lets Copilot retain architectural decisions across sessions, transparently in Git instead of a cloud black box. And the rubber duck agent, where a model from a different provider reviews the plan before the first line is written. Plus an honest look at custom agents, Bring Your Own Model with local LLMs, and what the new token-based billing means for all of this.
06
MCP Servers in Production: OAuth, Rate Limiting & Quotas
MCP is becoming the standard for connecting AI agents to external tools — but what happens when ChatGPT, Claude or thousands of autonomous agents start calling your API? This talk tackles the operational challenges MCP leaves unsolved: authentication, rate limiting and quotas, using OAuth for caller identification and Azure API Management for enforcement.
07
GitHub Copilot App: The Cockpit to Fly all Agents
The GitHub Copilot App is a desktop cockpit for coding agents: parallel sessions in isolated git worktrees, the complete PR lifecycle from issue to merge, and canvases where humans and agents work on shared state. In this talk, I show how to run multiple agent sessions in parallel without branch chaos, how to pick the right autonomy level from interactive to autopilot, and how Agent Merge handles CI fixes and the final merge. We build a canvas live from a single prompt and take a look at subagents, BYOK, and scheduled automations. The app is young and ships new features almost daily, so this is a snapshot from real usage: what works today, what is still rough, and how to get your first session running in five minutes.
Azure Meetup Bonn · 01.09.2026
🔗 Event
08
Where Did All My Credits Go? Understanding and Reducing Token Usage in GitHub Copilot
Since June 1, every Copilot interaction is billed by tokens, and the internet is full of tricks that promise 60, 70, even 95 percent savings. But which of them hold up? In this talk, I explain how tokens and AI credits actually work, show how to measure your own usage with tools you already have, and put several popular reduction techniques to the test on real tasks from my own repositories — with honest numbers instead of headline percentages. You'll leave knowing where your credits go and how to check any savings claim yourself.
09
Designing AI-Enabled .NET Applications with Microsoft Agent Framework
Adding AI to .NET applications is fairly easy today — for example, by calling large language models directly. The real challenge begins where AI is meant to be part of the application logic, not just a one-off helper. This talk introduces the Microsoft Agent Framework, which helps embed AI meaningfully beyond simple prompt calls. The focus: how to combine AI with existing .NET code, where to draw the line between deterministic logic and AI-driven behaviour, and which architectural considerations matter. Rather than treating AI as a replacement for established software-engineering principles, it looks at practical patterns, trade-offs and limitations of using AI in modern .NET applications — so you leave with a realistic sense of when the Microsoft Agent Framework is a sensible addition to your architecture and when simpler approaches are the better choice.
DWX 2026 · 30.06.2026
📑 Slides
10
GitHub Copilot Memory: From Black Box to Memory Bank
GitHub Copilot Memory promises to give your AI a long-term memory. But what does it actually remember, and who decides? In this talk, I explore the official Copilot Memory feature, a cloud-hosted black box that learns automatically but offers no user control, and contrast it with the Memory Bank pattern: a set of Git-tracked Markdown files that Copilot reads and updates with every task. No plugins, no external tools. Just a copilot-instructions.md and a folder of Markdown files that give your AI the context it needs to stop forgetting.
11
GitHub Copilot in Visual Studio: AI as a Pair Programmer
How can GitHub Copilot be used efficiently in a .NET environment? This session shows how developers can effectively integrate GitHub Copilot into their daily workflows in Visual Studio 2026. Matthias demonstrates step by step how GitHub Copilot supports writing, refactoring, testing, and documenting code, and how agentic coding with Agents and Instruction Files can be used productively. Participants will receive practical best practices for using GitHub Copilot consciously and effectively in their own .NET projects.
12
From Agents to Systems: Building MCP Servers for Practical AI Agents
AI agents become truly useful when they can interact with real systems rather than operate on text alone. This session focuses on how the Model Context Protocol (MCP) can connect AI agents to external tools and domain logic in a structured, maintainable way. It introduces the core concepts behind MCP and explains how an MCP server is designed, implemented, and exposed to AI agents. Using a practical example from a board game platform, the session shows how domain-specific functionality can be made available through well-defined tools and context, enabling meaningful interactions beyond simple prompt-based responses. The talk also covers architectural considerations such as hosting, the boundaries between agents and systems, and common pitfalls when integrating agents with existing applications.
13
Effiziente Entwicklung mit GitHub Copilot in Visual Studio (Efficient Development with GitHub Copilot in Visual Studio)
How can GitHub Copilot be used really efficiently in a .NET environment? This session shows how developers integrate GitHub Copilot into their daily workflows in Visual Studio 2026. Step by step, Matthias demonstrates how Copilot supports writing, refactoring, testing and documenting code, and how agentic coding with agents and instruction files can be used productively. Attendees get hands-on best practices for using GitHub Copilot consciously and effectively in their own .NET projects.
Technical Summit 2026, Munich · 28.01.2026
📑 Slides
14
Microsoft Session of AI: Develop your own MCP Server
This session covers the fundamentals of the Model Context Protocol (MCP) and why it plays a central role in the modern Microsoft AI ecosystem. We look at how an MCP server is structured, which components it needs and how requests and responses work. Then we build our own MCP server in .NET step by step — architecture, common implementation patterns and the most important API elements. Finally, I show how to test the server locally and plug it into your own AI workflows.
Azure Düsseldorf Meetup #43 · 13.01.2026
🔗 Event
15
AI-Driven Development with GitHub Copilot in Visual Studio
In this talk I show how I actually use GitHub Copilot in Visual Studio in everyday .NET work — beyond basic autocompletion. Many developers know Copilot by now, but few use it consistently and deliberately in their daily development process. I share how Copilot helps with typical tasks, how to collaborate with it sensibly, and how good prompts make a noticeable difference. Instead of theoretical concepts, it's about real-world experience, common pitfalls and concrete approaches to using Copilot more productively and purposefully.
.NET Dortmund User Group · 07.01.2026
🔗 Event
16
Azure Key Vault & Azure OpenAI: Secure AI Workflows in .NET
AI features are increasingly integrated into .NET applications, and every request to Azure OpenAI can contain sensitive data. This session shows how Azure Key Vault can manage credentials and configuration for Azure OpenAI cleanly and securely. In a live demo, a .NET application connected to Azure OpenAI is migrated from AppSettings secrets to Azure Key Vault. Attendees learn how managed identities, secret management and access control work together — and how to integrate AI features into existing applications responsibly and reliably.
Azure Cloud Native Cologne #20 · 16.12.2025
📑 Slides
More talks (2024–2025)
- 16.09.2025 Introduction to Azure Key Vault · Azure Düsseldorf Meetup #40
- 06.11.2024 Einführung in Azure Web Apps · Azure Cloud Native Cologne #10
// What I talk about
Topics & expertise
🤖 Agentic Coding & GitHub Copilot
Getting real value out of Copilot in Visual Studio & VS Code — prompting, instructions, skills, memory and agent mode for everyday .NET work.
🔌 Model Context Protocol (MCP)
Designing, hosting and securing MCP servers so AI agents can safely reach your APIs, databases and tools — from first demo to production.
🧩 Microsoft Agent Framework
Building AI-enabled .NET applications — agents vs. workflows, orchestration patterns, and knowing when deterministic code wins.
☁️ AI-enabled .NET on Azure
Bringing AI features into real ASP.NET Core applications — secure secrets with Key Vault, solid architecture, and the production decisions behind shipping GenAI on Azure.
// Beyond the stage
Community
Main organizer · since 2018
Azure Düsseldorf Meetup
I run the Azure Düsseldorf Meetup — a regular gathering for the local .NET, Azure and AI community to share talks, demos and hands-on sessions. Everyone's welcome.
azuredus.de → Organizer · 3rd edition in planningAgentic Coding Summit
I organize the Agentic Coding Summit — a recurring online conference for developers and architects who want practical, honest insight into agentic coding in real development work.
agentic-coding-summit.com → LinkedIn Newsletter📨 GitHub Copilot with Matthias
My newsletter on getting the most out of GitHub Copilot and agentic coding in the .NET world.
Subscribe at ghcp.liebeck.io →// Practice makes certified
GitHub certification practice exams
I create practical question-based courses to help you prepare for four GitHub certifications. Each course is available on Udemy with realistic practice exams and detailed answers.
GH-300 GitHub Copilot Certificate
Practice exams for the GitHub Copilot certification.
Free practice exams with voucher code
34452E07CDE1C01B7E8E through 13.11.2026.
If the code no longer works, please message me.
GH-200 GitHub Actions Certificate
Practice exams for the GitHub Actions certification.
Free practice exams with voucher code
AA1A4E930C76F876E818 through 13.11.2026.
If the code no longer works, please message me.
GH-900 GitHub Foundations
Practice exams for the GitHub Foundations certification.
Free practice exams with voucher code
054DE0F61726E7A912A3 through 13.11.2026.
If the code no longer works, please message me.
GH-600 GitHub Agentic AI Developer
Practice exams for the GitHub Agentic AI Developer certification.
Free practice exams with voucher code
C167ABC81A22497120C9 through 31.10.2026.
If the code no longer works, please message me.
// Things I've built
Projects
Gesellschaftsspieler-gesucht
I founded and run gesellschaftsspieler-gesucht.de — a social network that connects board-game players offline. With 20100+ members it's one of the largest German-speaking communities for board-game fans looking for people to play with, and I've built and operated it for over a decade.
- Built & operate the full stack — ASP.NET Core, Azure, CI/CD with GitHub Actions and SQL Server — for reliable, scalable operation
- Drive product strategy, community management and partnerships across the board-game scene
- Presented at SPIEL Essen 2013 & 2023, the world's largest board-game fair — and recently gave the platform its own MCP server
Open-source & research code
Argument mining, author profiling, semantic similarity and more — the open-source research tooling and experiments behind my NLP publications, on GitHub.
// Where I come from
Research background
Before moving fully into industry, I spent years in academic NLP research. It's the foundation for how I think about AI today.
PhD in Computer Science · Artificial Intelligence
Heinrich-Heine-Universität Düsseldorf · 2018
My dissertation, “Automated Discussion Analysis in Online Participation Projects,” applied text mining, argument mining and deep learning to large volumes of citizen feedback in German online participation platforms.
- M.Sc. & B.Sc. in Computer Science, plus a B.Sc. in Mathematics — HHU Düsseldorf
- 16+ peer-reviewed publications (ACL, SemEval & more)
- Argument & opinion mining, text mining, deep learning
- PAN @ CLEF 2016 — 2nd of 22 teams, 1st for English gender detection
// Hello
About me
I've been a professional software developer since 2009 and writing C# since 2006. After studying computer science and mathematics and earning my PhD in AI, I now work as a Senior Solution Architect at anyMOTION GmbH in Düsseldorf — leading client projects in ASP.NET Core, C# and Azure, where I introduced agentic coding to the team.
As a Microsoft MVP for Microsoft Foundry and Microsoft Azure, I organize the Azure Düsseldorf Meetup and speak regularly about GitHub Copilot, MCP and AI-enabled .NET. Off the keyboard I train Hyrox and play board games.
Experience
-
2021 — now
Senior Solution Architect
anyMOTION GmbH
- Tech lead on multiple client projects — own technical concepts, estimates and delivery
- Design web & cloud architectures; build ASP.NET Core, C# and EF Core solutions, test-driven
- Introduced agentic coding to the team; mentor developers and apprentices
-
2020 — now
malask GbR (side business)
- Founded the company behind SmartWarten, a SaaS platform for digital queues — built and run end-to-end
-
2019 — 2021
Senior Software Developer
anyMOTION GmbH
- Built web & API projects with ASP.NET Core, C#, EF Core and SQL Server (test-driven)
- Authored technical concepts and shaped system architectures; mentored apprentices
-
2014 — 2018
Researcher (PhD)
Heinrich-Heine-Universität Düsseldorf
- Researched NLP (argument & text mining) and published 16+ peer-reviewed papers
- Supervised 20+ Bachelor's, Master's and project theses
-
2009 — 2012
Junior Software Developer
istis Informations Systeme GmbH
- Developed ASP.NET / C# web applications as a working student
Awards, certifications & training
- 2026 GH-600 Developing in Agentic AI Systems
- 2026 Microsoft MVP — Microsoft Foundry & Microsoft Azure
- 2025 GH-300 GitHub Copilot
- 2025 GitHub Foundations
- 2024 DP-600 Fabric Analytics Engineer Associate
- 2024 AZ-305 Azure Solutions Architect Expert
- 2022 SC-900 Security, Compliance & Identity Fundamentals
- 2022 PL-900 Power Platform Fundamentals
- 2022 MB-920 Dynamics 365 Fundamentals (ERP)
- 2022 MB-910 Dynamics 365 Fundamentals (CRM)
- 2020 DP-900 Azure Data Fundamentals
- 2020 AI-900 Azure AI Fundamentals
- 2020 AZ-104 Azure Administrator
- 2020 Professional Scrum Master (PSM I)
- 2019 AZ-203 Developing Solutions for Azure
- 2019 AZ-900 Azure Fundamentals
- 2019 DP-100 Azure Data Scientist Associate
- 2019 AEVO — Certified Trainer
Let's talk about Agentic Coding
Looking for a speaker for your conference, meetup or team? Or want to compare notes on Copilot, MCP and AI-enabled .NET? The best way to reach me is a quick message on LinkedIn.
Impressum
Information pursuant to § 5 DDG (German Digital Services Act).
Service provider
Dr. Matthias LiebeckVölklinger Straße 2f
40219 Düsseldorf, Germany
Contact
Email: webmaster@liebeck.ioPhone: +49 179 4774320
Privacy Policy
This is a static information site with no contact forms, tracking or advertising cookies. Fonts are served locally; no external font or CDN services are used.
Controller
Dr. Matthias Liebeck · Völklinger Straße 2f, 40219 Düsseldorf, Germany · webmaster@liebeck.ioHosting & server logs
This site is hosted on Microsoft Azure App Service (Microsoft Corporation; EU region “West Europe”). When you access the site, the host processes technically necessary data (IP address, date and time of access, requested file, data volume, browser type) in server log files. Legal basis: Art. 6 (1) (f) GDPR (legitimate interest in secure, stable operation). See the Microsoft Privacy Statement.
Your rights
You have the right to access, rectification, erasure, restriction of processing, data portability and objection, as well as the right to lodge a complaint with a supervisory authority.
Last updated: June 2026