/Events /AgentCon 2025 - Berlin /Schedule
AgentCon 2025 - Berlin
Tuesday, June 24
Checkin Attendees
Plenary ·Welcome and Opening – Global AI Community and CODE University of Applied Sciences
Join us kicking off and celebrate together AgentCon Berlin
Prof. Dr. Peter Ruppel is a Professor of Software Engineering and was Vice President of CODE University of Applied Sciences from 2021 to 2023. He completed his studies in computer science and his doctoral thesis at Ludwig-Maximilians-Universität in Munich.
As the Chief Executive Officer of CODE University of Applied Sciences, I lead an innovative and international institution that empowers digital pioneers to create meaningful solutions for the future. With a Ph.D. in Economics from the University of St. Gallen, I have a strong academic background and a passion for lifelong learning. I have over 10 years of experience in leading comprehensive transformation projects in the education sector, working with top management and private equity clients. I was the CEO of Macromedia University, the leading private onsite and hybrid university for the digital and creative industries in Germany, and the Managing Director of Galileo Global Education Germany, where I prepared the organization to serve multiple brands. I have a proven track record of building and growing successful education businesses, leveraging my expertise in strategy, development, and management consulting.
I am an AI, XR expert and Microsoft MVP. I spoke at TED Conferences, Microsoft Build / Ignite, WeAreDevelopers World Congress, focusing on innovation and solving organizational challenges with advanced technology. My projects, which make complex tech accessible and engaging, span sectors like Cultural Heritage, Healthcare, and Creative Coding. As a multiplier for immersive technologies and AI, I am committed to fostering collaborative growth and learning worldwide
Christian Glessner is a globally recognized technology expert and founder of Hololux, a leading innovation agency specializing in AI and mixed reality solutions for enterprise transformation. As a 15-time Microsoft MVP and recipient of the 2024 German Innovation Award, he has spent over two decades helping organizations worldwide leverage cutting-edge technologies to drive digital innovation and operational excellence. Christian's expertise in AI agents, immersive technologies, and enterprise solutions has established him as a prominent speaker at major international conferences. Known for his engaging presentations that transform complex AI concepts into practical, implementable strategies, he empowers business leaders to navigate the digital transformation landscape with confidence. His work bridges the gap between emerging technology and real-world business applications, delivering measurable impact for organizations ready to embrace the future of digital innovation.
Keynote: Building the Future of Agents
When everything is an agent, what is an agent? - lets break it down to start the day. Join us as we explore an agent scenario for Contoso Pizza, breaking down each component from models and prompts to tools. Engage in audience participation to share your AI experiences. Let's demystify agents and continue connecting on this exciting topic beyond the event
Amy Boyd, with a background in Computer Science and a focus on Natural Language Processing and Machine Learning, is a Principal Cloud Advocate at Microsoft. Her passion for AI and Machine Learning has led her to work on global projects, engage with developer communities, and devise strategies for AI Developer community engagement. In her role, Amy not only creates and delivers content but also shapes the product experience for developers, drawing on her extensive experience in the field.
Skip the Vocabulary: Building Language Models That Read Character Patterns Directly
What if AI models could understand text without needing vocabularies at all? This isn't a hypothetical future—it's happening now with a breakthrough architectural approach that eliminates the entire tokenization layer and lets models read character patterns directly.
Archana is a Developer's Advocate at Aleph Alpha. Prior to this, she worked as a Program Manager and Community Leader in various AI Safety and women in tech communities. She has over 8 years of experience as an AI engineer. As a Board Director at Women in Machine Learning and a fellow at Python Software Foundation, she champions responsible AI development through community-driven initiatives.
Data Security meets AI
How to protect your data and leverage the power of artificial intelligence
In this speaker slot, you will learn about the challenges and opportunities of data security in the age of AI. You will discover how to safeguard your data from unauthorized access, manipulation, and theft, while also harnessing the potential of AI to enhance your data analysis, insights, and decision making. You will also hear about the best practices and tools for data security and AI, as well as the ethical and legal implications of using AI on sensitive data.
This speaker slot is suitable for anyone who works with data and wants to learn more about data security and AI. Whether you are a data analyst, a data scientist, a data engineer, a data manager, or a data enthusiast, you will find this speaker slot informative and engaging.
You will also have the opportunity to ask questions and interact with the speaker and other participants.
RAPHAEL KÖLLNER Compliance Manager Raphael Koellner combining the law and the information technology in particular the cloud computing. He worked for universities, Microsoft Germany (DX) and Microsoft Partner as consultant. As CEO for KöllnService GmbH he works for huge international companies to migrate and implementate cloud solutions on a technical and a contract/law way. Since 2019 Raphael Köllner is Microsoft Regional Director. He is the community guy, MVP for Office Server & Services, Windows 10 Insider MVP, MCT, Microsoft Student Partner Evangelist and become an attorney in Europe. He is the lead of the Office 365 Usergroup Germany, the Office 365 DACH, Leader of the Office 365 conference Germany, the SharePoint Saturday Cologne and one of 30 Windows 10 Top Insiders. Moreover he is member of the TEAMS Elite Team and the organizer of the TeamsCommunityDay. In this connection he leads the biggest Windows 10 Insider Community in Germany. MVP: https://mvp.microsoft.com/en-us/mvp/Raphael%20%20Koellner-5000185 Microsoft Regional Director: https://rd.microsoft.com/en-us/raphael-koellner
Scaling AI Agents Efficiently - Token Optimization Techniques with ScaleDown
This hands-on technical workshop teaches AI professionals and developers how to optimize token usage in AI agent systems using the ScaleDown Open-Source Python package. When building and deploying AI agents, participants will learn practical techniques to reduce token consumption, improve response times, and lower API costs.
Workshop Outline
Welcome & Setup (10 min) I'll introduce myself, explain the token efficiency problem, and make sure everyone's ready to follow along with the code examples. Token Optimization Essentials (15 min) We'll explore how ScaleDown works and look at its core components for analyzing and optimizing prompts. I'll demonstrate key techniques with before-and-after examples that show dramatic token reduction. Real-World Agent Use Cases (35 min) This is the heart of our workshop - we'll dive into implementing optimization for these practical scenarios: Multi-Turn Conversations: Compress conversation history without losing context while reducing tokens by 60% Document Processing: Strip boilerplate text from legal and technical documents before analysis, cutting usage by 70-80% Multi-Agent Communication: Streamline handoffs between specialized agents to eliminate redundant context and improve system efficiency Meeting Transcripts: Transform verbose transcripts into compact summaries by removing filler words and redundant speaker identifications
Q&A and Next Steps (10 min)
Bring your questions! Here is the Python package we will be using during the workshop: https://github.com/scaledown-team/scaledown-package
While we've focused on cost and performance benefits today, it's worth noting that these optimizations also reduce the carbon footprint of your AI systems. By implementing ScaleDown, you're not just saving money and improving performance – you're also contributing to more sustainable AI that consumes less energy with every interaction.
Soham Chatterjee is a Senior AI Engineer at Rasa and creator of the ScaleDown open-source Python package. With over 8 years of ML experience, including his role as AI/ML Lead at Sleek, he has pioneered token compression techniques that reduce AI costs by up to 80%. Soham's work on the ScaleDown API focuses on both cost optimization and sustainability, having conducted extensive research on reducing the carbon footprint of AI systems through intelligent prompt optimization. His expertise spans production ML systems, LLM optimization, and sustainable AI practices.
Power your AI Transformation with the Microsoft Cloud
Lets discuss the rapid advancements and future potential of Artificial Intelligence and the stages where organizations typically get stuck in AI implementation, from proof of concept to AI-led business transformation We will talk about the unique capabilities of AI agents, such as reasoning, acting, and learning, and showcases various use cases, including customer support and engagement. Additionally, we will explore Microsoft's approach to building AI agents, the role of Azure AI Foundry, and the importance of responsible AI practices
See https://www.linkedin.com/in/marianomailos/
From Vision to Corporate: Embedding Agentic AI Across a 600-Person Enterprise
In early 2025, we launched a company-wide initiative to embed agentic AI into the DNA of S-Communication Services — a hybrid tech organization of 600 people within the Sparkassen Finanzgruppe. In this talk, I’ll share how we approached the challenge of rolling out agentic AI across diverse departments — from marketing and product to software development — and what we learned along the way. Expect real use cases, honest lessons from implementation friction, and insights into what it takes to scale agentic AI across enterprise systems and culture. If you're interested in AI beyond the hype talks, this session is for you.
I’m a Banking Technologist, University Lecturer, and AI Enthusiast driven to bridge technology, leadership, and innovation. Currently running an Innovation Team, I oversee strategic transformation projects and guide cross-functional teams to unlock AI’s potential. My background in the German Armed Forces shaped my resilience, strategic foresight, and cross-cultural collaboration skills. Today, I focus on the intersection of technology and society, teaching at CODE University and expanding my AI expertise at The University of Texas at Austin.
Break
TetRis ·Achieving Precision in AI: Retrieving the Right Data Using AI Agents
In the race to harness the power of generative AI, organizations are discovering a hidden challenge: precision. Models are only as effective as the data they access, yet most approaches to Retrieval-Augmented Generation (RAG) lack the dedicated, fine-tuned pipelines needed to ensure the right information is delivered at the right time.
Today, most RAG systems pull from vast, generalized data lakes, leading to noisy outputs and frustrating inefficiencies. The result? Wasted resources, inconsistent responses, and missed opportunities for real-time decision-making. But what if you could create an AI system that doesn’t just retrieve data—but understands its context, delivering precise, actionable insights in milliseconds?
This is where agenticRAG comes into play—a breakthrough in AI architecture that pairs dedicated retrieval pipelines with intelligent agents to deliver pinpoint accuracy. By segmenting your data storage and retrieval processes specifically for training vs. inference, you can achieve hyper-focused precision while dramatically reducing latency and costs. Imagine an AI system that knows exactly what data it needs and how to get it with zero lag—a system that’s tuned to perform like a well-trained expert in your domain.
Curious to discover how you can optimize your AI applications for laser-focused accuracy? Join me as to learn more about AgenticRAG and fine tuning your models.
Adi is an experienced Software Engineer and people manager. For most of her professional life, she has worked with data and machine learning for operations and analytics. As a data practitioner, she developed algorithms to solve real-world problems using machine learning techniques and leveraging expertise in Apache Spark, Kafka, HDFS, and distributed large-scale systems. Adi has taught Spark to thousands of students and is the author of the successful book — Scaling Machine Learning with Spark. In early 2024, she embarked on a new adventure with data streaming, specifically Flink, and she can't get enough of it.
AI in the Boardroom: Cultural, ethical, and sustainability research issues
While developers and creators are racing ahead with exciting technical developments in AI, the academic research community is also exploring challenging social and ethical issues that these tools reveal. Join faculty and graduate students from ESCP Business School Berlin campus for a lively cross-disciplinary conversation and exploration of these topics.
Currently working at ESCP Business School, Carla's career spans diplomacy, counseling, and international teaching and training. She is committed to the development of critical thinking and compassion action - Always exploring ways to bring emerging technologies to her students
Raga Teja Sudhams Kanaparthi is a PhD candidate and Research Associate at ESCP Business School Berlin, where his work focuses on responsible management, with a particular emphasis on artificial intelligence (AI) and digital stakeholder communities. With a diverse academic background, he holds a Master of Research in Management from the University of Toulouse Capitole 1 and a Master of Science in Marketing from CIHEAM Zaragoza, Spain. Raga's professional journey spans five years as a marketing manager across various global contexts and as a researcher at the Social and Innovation Marketing Lab at Toulouse Business School. His cross-disciplinary expertise allows him to explore how AI can not only be governed ethically and sustainably but also be leveraged as a tool for promoting sustainability, responsibility, and ethical practices. He believes that the future of technology depends not just on innovation, but on thoughtful integration of ethical, responsibility and sustainability values, and he is committed to helping shape that future.
Zhengtian Xia is a doctoral researcher at ESCP Business School, Berlin Campus. He is affiliated to the Chair of International Management and Strategic Management. His research interest covers international strategy and artificial intelligence.
Herding AI Cats: Semantic Kernel Multi-Agents Scenarios.
Semantic Kernel: Beyond the "Hello, World." This deep dive gets elbows-deep into the framework for serious LLM, plugin, and agent orchestration. We'll tear down a multi-agent software dev scenario to expose advanced techniques in orchestration, planning, and integration. Expect real code, real problems, and how to beat Semantic Kernel into submission for complex tasks. If you want the advanced stuff, this is it.
Marc Plogas discovered his passion for technology at the age of six when an Atari sparked his interest in programming. He holds a Master's in Computer Science and spent nearly a decade freelancing as a software, mobile, and test engineer, as well as a solution architect, before joining Microsoft. At Microsoft, he played an instrumental role in pioneering technologies such as Windows Mixed Reality, Desktop Bridge, and WinUI 3, and later advised startups worldwide on optimizing their cloud architectures with a focus on reliability, scalability, cost efficiency, and security. Now embarking on a new chapter, Marc eagerly shares his extensive expertise in IoT, AI, ML, and Mixed Reality through engaging public talks and workshops.
Boost Developer Efficiency with Al Agents Through an Open Agentic System
The Agentic system is an open platform designed to help developers integrate Al agents into their projects seamlessly. Developers can leverage pre-existing Al agents or build their own, automating tasks such as documentation, test case creation, and code quality assessment to boost efficiency.
Our primary focus is to provide a scalable, extensible Al-driven automation solution. Instead of emphasizing individual Al agents, we offer a platform-first approach that integrates effortlessly with widely used project and code management tools like GitHub. This enables users to incorporate Al agents with minimal effort, streamlining repetitive development tasks.
By standardizing Al agent integration, developers can reduce manual workload, improve productivity, and accelerate delivery cycles. Additionally, businesses benefit by optimizing workforce efficiency and reducing project costs through Al-powered automation.
Senior Software Engineer & AI-Driven Development Specialist With over 20 years in the software development sector, Daniel stands out as a seasoned Software Engineer and Test Automation aficionado. A proud alumnus of the Technical University Braunschweig with a Master’s in Computer Science, he has donned multiple hats – from Senior Java Developer to Senior Fullstack Developer, and CTO. What truly distinguishes Daniel is his expertise in AI-powered software development. He’s been at the forefront of integrating AI into the software development lifecycle, guiding clients on leveraging cutting-edge AI solutions like GitHub Copilot, Amazon Q, Windsurf, and tailored AI agents for enhanced test generation, code reviews, story writing, and AIOps. Daniel is also a key architect behind customized build GenAI agents and build multi-agent system designed to autonomously handle complex development tasks. These agents collaborate across various stages of the SDLC, from planning and coding to testing and releasing, significantly boosting development speed, quality, and scalability.
The AI Advantage: Unlocking Agent Potential through Smart Prompting
AI is transforming software development! Those who welcome this change and work hand-in-hand with their AI teammates will thrive in this new landscape. Developers must acquire the skills to craft effective prompts, test and optimize agents, and understand their limitations. In this interactive session, Sven will explore the art of AI prompting, showcase multi-agent workflows, and share some handy tips to help you unlock the full potential of these agents. While we'll be utilizing Atlassian’s AI platform, Rovo, the insights you gather here will apply to any prompt-based AI agent. Join this session to enhance your AI usage capabilities and elevate your developer experience—one agent at a time.
Sven Peters, DevOps advocate at Atlassian, has been studying trends in software development for the last 15 years uncovering the cultural and technical attributes to help development teams work effectively and drive innovation. He has 20 years experience in writing code, leading teams, and sharing his experience with thousands of developers at uncountable conferences in 25+ countries.
Break
TetRis ·Boost your coding productivity with Github Copilot Agent
This year, Github introduced the agent mode for GitHub Copilot in Visual Studio Code, announcing the general availability of Copilot Edits, and providing a first look at our SWE agent. In this talk, we will demonstrate the difference between the three modes and show the advantages to use the agent mode. In agent mode, Copilot will iterate on not just its own output, but the result of that output. And it will iterate until it has completed all the subtasks required to complete your prompt. Instead of performing just the task you requested, Copilot now has the ability to infer additional tasks that were not specified but are also necessary for the primary request to work.
Alexander Wachtel has been recognized as a Microsoft AI MVP since 2020 for his community work in AI area. During his Ph.D. research at the Karlsruhe Institute of Technology (KIT), he has published several AI papers in the area of Human-Computer-Interaction, Natural Language Processing & Cognitive Services. Furthermore, he is still teaching agile software engineering at the KIT using Azure DevOps. Due to his teaching activities at the KIT, he was also awarded as a Microsoft Innovative Educator (MIE) expert from 2016 to 2019. Due to his many years of experience, he has been also honored as a Microsoft Certified Trainer (MCT) since 2013. He provides Microsoft consulting for the industry and public sector, and teaches classes for Microsoft Certified Azure Solutions Architect Expert and Microsoft 365 Certified Administrator Expert.
Julia is a Global Black Belt for Developer Productivity at Microsoft, helping large-scale organizations accelerate dev workflows through pragmatic, end-to-end platform engineering. With a background as a Software Engineer, Dev Lead, and Cloud Solution Architect, she speaks fluent developer and management—and knows how to bridge the gap without losing either side in translation. She’s a big-picture thinker who also sweats the details. Web and frontend will always have a place in her ❤️. Outside work, Julia is a gamer, a proud fahrradmaus, and a former quarterback for a women’s American football team (yes, really). On Wednesdays, she smashes the patriarchy.
Building a Multi-Agent Financial Assistant with Amazon Bedrock
This session introduces a powerful financial assistant that uses multiple AI agents built with Amazon Bedrock. We’ll show how different agents—each specializing in areas like historical data, pricing, risk assessment, market sentiment, and portfolio management—work together to provide well-rounded financial advice. We’ll explain why it’s important for these agents to interact and share information, leading to more accurate and useful insights. Through examples, we’ll demonstrate how these agents cooperate, handle disagreements, and combine their knowledge to make smart recommendations. Attendees will learn how to create and manage these agents using AWS tools, with a focus on Amazon Bedrock. By the end of the session, you’ll understand how to use multi-agent systems to build advanced AI-powered financial tools that can handle complex market situations and offer more reliable, tailored financial guidance.
Liza (Elizaveta) Zinovyeva is an Applied Scientist at AWS Generative AI Innovation Center and is based in Berlin. She helps customers across different industries to integrate Generative AI into their existing applications and workflows. She is passionate about AI/ML, finance and software security topics. In her spare time, she enjoys spending time with her family, sports, learning new technologies, and table quizzes.
Aiham Taleb, PhD, is a Senior Applied Scientist at the Generative AI Innovation Center, working directly with AWS enterprise customers to leverage Generative AI across several high-impact use cases. Aiham has a PhD in unsupervised representation learning, and has industry experience that spans across various machine learning applications, including computer vision, natural language processing, and medical imaging.
The Hidden Patterns of Agentic AI—How Context Shapes Intelligence
AI agents don’t think in isolation—they follow patterns of behavior based on the context they retrieve, the decisions they make, and the data they rely on. Yet most AI today is blind to its own patterns, reacting without understanding the broader structures influencing its actions. This workshop will uncover the emergent patterns that shape agentic AI and how we can design agents that leverage real-time context to become more adaptive, reliable, and effective. We’ll explore: Common behavioral patterns in AI agents—how they retrieve, process, and act on context. Why agentic AI needs structured, evolving context to break free from rigid, static behavior. How to analyze, refine, and optimize agentic patterns for better decision-making and adaptability. Through hands-on exercises, participants will learn how to trace, visualize, and shape AI agent behavior using dynamic context retrieval, interaction modeling, and live data adaptation. If we want AI agents to be truly intelligent, we need to understand the patterns they follow, the context they need, and the behaviors they evolve—this workshop will teach you how.
Michael Hunger has been passionate about software development for more than 30 years. For the last 15 years, he has been working on the open source Neo4j graph database filling many roles, most recently heading product innovation and GenAI. As a developer Michael enjoys many aspects of software development and architecture, learning new things every day, participating in exciting and ambitious open source projects and contributing and writing software related books and articles. Michael spoke at numerous conferences and helped organize others. Michael helps kids to learn to program by running weekly girls-only coding classes at local schools.
D Insight - A system for digital sales of banking products using AI agents
The presentation showcases D-Insight, an AI-powered system for personalized digital sales in banking. It uses transaction-based behavioral triggers to deliver timely, relevant product offers, boosting conversion rates and user satisfaction. The system supports both cloud and on-prem deployment and operates autonomously.
Software Engineer and University Professor with a Ph.D. in E-Business and over 10 years of experience, specializing in full-stack development with PHP (Laravel, Symfony), JavaScript, Node.js, React, and Next.js. Actively involved in AI system architecture and deployment, including LLM integration, distributed inference, and intelligent agent workflows. Skilled in AWS and DevOps, with a strong background in leading development teams to deliver scalable APIs and AI-powered applications. As a professor, teach courses in software development and mentor students on academic and real-world projects. Excellent communicator with strong presentation skills and a talent for building lasting relationships with colleagues, clients, and partners.
Vibe Coding: Building Voice AI Agents with n8n and Vapi
In this talk, Aemal Sayer shares how AI agents are reshaping software development and how voice is becoming a primary interface for interacting with computers. He introduces the Vibe Coding methodology, an intuitive, fast-paced way to build with tools like Cursor, Task Master, n8n and Vapi. With 23 years of hands-on experience, a background in AI and computer science, and five companies under his belt, Aemal offers hard-earned insights into building modern products with emerging AI technologies.
Aemal Sayer is a seasoned software engineer and tech entrepreneur with over 23 years of experience in software development and 8 years in AI applications. Specializing in building AI agents for SMEs in Germany, he focuses on automating financial processes such as bookkeeping and invoice management, ensuring compliance with EU regulations like the AI Act and GDPR. Throughout his career, Aemal has collaborated with industry leaders including Klarna, Siemens, Allianz and more, delivering transformative AI solutions that optimize operations and drive measurable results.
Integrating AI Agents into Agile Teams: Collaboration or Replacement?
The proliferation of AI agents within tech teams has brought the question of agency in collaboration back into the focus. How do we distribute and manage work so that our human team members can thrive? Is a "people over processes" paradigm in danger when processes themselves get more agency? And can our AI agents actually join our retro?
Prof. Dr. Peter Ruppel is a Professor of Software Engineering and was Vice President of CODE University of Applied Sciences from 2021 to 2023. He completed his studies in computer science and his doctoral thesis at Ludwig-Maximilians-Universität in Munich.
Amy Boyd, with a background in Computer Science and a focus on Natural Language Processing and Machine Learning, is a Principal Cloud Advocate at Microsoft. Her passion for AI and Machine Learning has led her to work on global projects, engage with developer communities, and devise strategies for AI Developer community engagement. In her role, Amy not only creates and delivers content but also shapes the product experience for developers, drawing on her extensive experience in the field.
Adi is an experienced Software Engineer and people manager. For most of her professional life, she has worked with data and machine learning for operations and analytics. As a data practitioner, she developed algorithms to solve real-world problems using machine learning techniques and leveraging expertise in Apache Spark, Kafka, HDFS, and distributed large-scale systems. Adi has taught Spark to thousands of students and is the author of the successful book — Scaling Machine Learning with Spark. In early 2024, she embarked on a new adventure with data streaming, specifically Flink, and she can't get enough of it.
Creative Blogger Multi-Agent with Azure
In this session, we will explore how to create a multi-agent system to help you write articles for your blog. We will use various agents to perform different tasks such as researching and gathering information on the topic, conceptually summarizing the information, and drafting a post that provides a solid foundation for our blog post.
14 años de experiencia en el mundo .NET, diseñando y construyendo soluciones basadas en Azure. Desde hace 2 años, enfocado más al mundo Power Platform y la Inteligencia Artifical.