About
Hi, I'm Johannes — a computer scientist with a Master's degree and a strong focus on Artificial Intelligence. My research centers around Reinforcement Learning, Deep Learning, and autonomous systems, areas where I'm deeply invested both academically and practically. In addition to AI, I bring experience in web development and working with medical sensors.
Outside of work, I enjoy playing badminton, making music on the piano, and exploring the world through travel. I'm always curious, always learning, and always looking for meaningful challenges to solve.
Specializations
Projects
Diverse and Adaptive Behavior Curriculum for Autonomous Driving: A Student-Teacher Framework with Multi-Agent RL
Developed a novel reinforcement learning framework for training autonomous vehicles using adaptive curriculum learning. Designed a multi-agent system where an AI-controlled "teacher" dynamically generates diverse and challenging traffic scenarios to improve the robustness of a self-driving "student" agent. Evaluated the approach in CARLA, demonstrating improved adaptability over rule-based traffic training. The work was created in my Master Thesis at FZI Research Center for Information Technology in Karlsruhe and resulted in a publication, which was accepted at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025).
Intelligent Actor Control for Optical Sorting Using Deep Reinforcement Learning
Developed and evaluated intelligent control strategies for an energy-efficient optical sorting system using multiple mechanical actors. Initially implemented and tested online planning methods such as POMCP and heuristic approaches to establish performance baselines. Designed both discrete- and continuous-time simulation environments and applied state-of-the-art deep reinforcement learning algorithms (PPO and SAC) to learn optimal control policies offline. Achieved significant improvements in both accuracy and real-time capability compared to prior methods, demonstrating the potential of deep RL for real-world industrial sorting tasks.
EEG Measurement in the Context of Flow and Immersion
Flow and immersion describe mental states where individuals become deeply engaged in an activity—commonly known as "being in the zone." Video games are particularly effective at inducing these states, but traditional self-assessment methods rely on subjective, delayed responses. At Fraunhofer IOSB, I contributed to an experiment that aims to objectively measure flow and immersion using physiological data. My work introduced EEG-based brain activity tracking, integrating new hardware and software into the existing experiment setup. A proof-of-concept study confirmed correlations between increased theta and beta wave activity and flow states, aligning with prior research.
Art Project: Infinite Poem
Collaborated with an artist to create an interactive art installation that generates an endless poem by randomly combining short sentence fragments. The verses were automatically typed on a vintage Brother CE-50 typewriter while a prerecorded voice recited the lines in sync, blending mechanical, textual, and auditory expression.
Freelance Web Development
One of my first side projects in my computer science career, where I designed and built websites for clients such as a local football club. Used CMS platforms like WordPress together with custom frontend development in HTML, CSS, JavaScript, and related frameworks to deliver tailored, user-friendly solutions.
Get In Touch
I'm always interested in discussing new opportunities, collaborations, or just talking about the latest developments in AI and autonomous systems. Feel free to reach out!