Robot Learning Engineer

ABOUT THE ROLE This is an entry-level research engineering role at an early-stage industrial robotics startup, sitting at the intersection of machine learning and real factory hardware. You'll apply state-of-the-art perception, reinforcement learning, and imitation learning to a physical robotic work cell built to automate demanding industrial tasks — surface finishing, welding, and coating. It's a rare opportunity for a hungry new grad to ship ML research directly onto production robots. WHAT YOU'LL DO - Research and evaluate ML models for robot perception and task understanding. - Apply computer vision and deep learning to multi-modal sensor data, including cameras, depth sensors, and force/torque inputs. - Design and run experiments with reinforcement learning and imitation learning for robot control. - Integrate trained AI models into a ROS 2–based robotics software stack. - Iterate rapidly on experiments and translate research findings into real-world factory deployments. WHAT WE'RE LOOKING FOR - BSc or MSc in Robotics, Computer Science, AI/ML, or a related field — or equivalent hands-on experience. - 0–3 years of experience; strong fundamentals matter more than seniority. - Proficiency in Python and experience with PyTorch or TensorFlow. - Exposure to computer vision and/or robot learning (thesis, internship, open-source, or research counts). - Familiarity with robotics simulation environments (e.g. Isaac Sim, MuJoCo, PyBullet) is a plus. - Experience with sim-to-real transfer, 3D perception, or point clouds is a bonus. - Research publications or coursework in RL or imitation learning are welcome signals. - Eligible to work in Germany and able to work on-site in Munich — no visa sponsorship available. LOCATION On-site in Munich, Bavaria, Germany. This role is fully on-site; remote work is not available. Candidates must already be eligible to work in Germany — visa sponsorship is not offered.