Public summary
An international high-tech company in Munich is seeking a Working Student to support AI research by developing models that improve the realism of simulated sensor and actuator data to better match real-world robotic tasks. The role involves working with physics simulators, applying domain randomization techniques, and collaborating with research teams to enhance model training. Candidates should be enrolled in a relevant Bachelor’s or Master’s program and proficient in Python, with familiarity in robotics simulation and machine learning concepts.
Location and work setup
- Location
- Munich
- Remote status
- On-site
- German requirement signal
- No German Required Detected
- Detected job language
- English
Responsibilities
Contribute to developing learned transformations that map simulated sensor and actuator outputs closer to real-world distributions; evaluate performance of models trained on synthetic data for real robot tasks; support reproducible and production-grade simulation pipelines; apply domain randomization and augmentation to reduce simulation-to-reality gaps; coordinate with research teams to identify critical realism gaps affecting model training.
Qualifications
Enrollment in a Bachelor's or Master's degree program in Computer Science, Robotics, Machine Learning, or related discipline; proficiency in Python programming; working knowledge of physics simulators such as MuJoCo, Isaac Sim, or Gazebo; foundational understanding of model evaluation for sim-to-real transfer. Beneficial skills include exposure to sensor noise modeling, actuator dynamics, domain randomization, experience with ROS or ROS2, and awareness of versioning and reproducible pipeline design.