Machine Learning Engineer specializing in bridging the gap between deep learning research and safety-critical automotive deployment.
Note: My public GitHub contribution graph looks a bit quiet because my daily work happens behind the closed doors of a major German automotive manufacturer.
🚗💨 I solemnly swear that my GitHub Enterprise graph is a lush, vibrant forest of green with well over 1,000 contributions this year. Trust me, the code is moving!
- 🏎️ Current Role: Machine Learning Engineer at BMW, bringing AI-based automated driving to production.
- 🔬 Past Life: Research Engineer at Bosch Research, working on the future of connected mobility and AI-based radar perception.
- 🎓 Education: Ph.D. in Computer Engineering from Ulm University and M.Sc. from RWTH Aachen University.
- 🎯 Current Mindset:
Research [▬▬▬▬▬▬▬▬▬▬▬▬⚡️▬▬▬] Production(roughly 80% shipping code, 20% reading papers)
| Domain | Focus areas |
|---|---|
| Machine Learning | End-to-End Automated Driving, Data Curation, Model Optimization, Deployment |
| Embedded & Systems | Safety-Critical Software Architecture, Modern Embedded C++, Vehicle Testing |
| Automotive Safety | ISO 26262, Dependability, MISRA C++ |
- Real-World Evaluation of two Cooperative Intersection Management Approaches Marvin Klimke, Max Bastian Mertens, Benjamin Völz, Michael Buchholz — IEEE Intelligent Transportation Systems Magazine, 2026. DOI / arXiv
- Automatic Intersection Management in Mixed Traffic Using Reinforcement Learning and Graph Neural Networks Marvin Klimke, Benjamin Völz, Michael Buchholz — IEEE Intelligent Vehicles Symposium (IV), 2023. DOI / arXiv
- Predicting the Time Until a Vehicle Changes the Lane Using LSTM-Based Recurrent Neural Networks Florian Wirthmüller, Marvin Klimke, Julian Schlechtriemen, Jochen Hipp, Manfred Reichert — IEEE Robotics and Automation Letters, 2021. DOI / arXiv
- 🛠️ Deep Learning to Production: Bringing deep learning models for end-to-end automated driving to series automotive hardware.
- 🤝 Team Enabling & Collaboration: Acting as a bridge between research teams, product owners, and embedded developers. I actively drive knowledge sharing, moderate technical alignments, and streamline interfaces across diverse stakeholders to keep engineering moving fast.
- 🔒 Safety First: Ensuring that Claude, GPT, and friends stay in the passenger seat and never compromise on our safety goals.
- 🚴 Off the Grid: Outside of IDEs, you can usually find me cycling on the weekend or pushing my limits on a running trail.
- 🔢 Erdős Number: 4 (via my co-authors during my time at Mercedes Benz).
- ⌚ Garmin Telemetry: Currently averaging 10,632 steps/day and climbing roughly 194 floors per week.
- ⛷️ Downhill Skiing Vmax: 85.8 km/h (in the Austrian Alps)
Building AI that doesn’t just run in the cloud, but actually hits the road.