Together, we are taking the step into a new era of cognitive robots:
As an Expert in Reinforcement Learning and Behavior Cloning for Humanoid Robots at NEURA Robotics, you will lead the development and implementation of cutting-edge algorithms and systems. You will collaborate closely with our robotics engineering team to enhance the autonomy, adaptability, and interaction capabilities of our humanoid robots. This includes:
Improve the learning pipeline for robust locomotion and loco-manipulation on the 4NE1 humanoid, from policy training to validated on-robot deployment
Explore and adapt cutting-edge methods (behavior foundation models, general motion tracking, teleoperation) and ship to customers and other teams.
Drive sim-to-real transfer (domain randomization, actuator modeling, physics alignment) to deploy policies from simulation onto real hardware with minimal performance gap.
Integrate reinforcement learning and behavior cloning approaches to enable robots to learn from both simulated and real-world interactions.
Collaborate with controls, hardware, and perception teams to ensure seamless integration of algorithms with robot hardware and sensors.
Conduct experiments and simulations to evaluate and validate algorithmic performance in various scenarios.
Master’s or Ph.D. in Robotics, Computer Science, or a related field
Proficiency in programming languages like Python, C++ with hands-on experience in robotics frameworks (e.g., ROS2) and robotic hardware platforms
Experience in developing and deploying reinforcement learning algorithms and behavior cloning techniques for robotic systems, with a strong grasp of humanoid robot dynamics, control theory, and sensor integration.
Hands-on experience with Isaac Lab / Isaac Sim and/or MuJoCo, and PyTorch.
Strong analytical skills and a proven ability to deliver innovative solutions in robotics, ideally including a track record of publications or contributions to conferences and journals
Your communication skills make you shine
You have a perfect command of the English language and, best of all, speak German well
Skills Required
- Master's or Ph.D. in Robotics, Computer Science, or a related field
- Proficiency in Python and C++
- Hands-on experience with robotics frameworks (e.g., ROS2) and robotic hardware platforms
- Experience developing and deploying reinforcement learning and behavior cloning for robotic systems
- Strong understanding of humanoid robot dynamics, control theory, and sensor integration
- Hands-on experience with Isaac Lab / Isaac Sim and/or MuJoCo, and PyTorch
- Experience with sim-to-real transfer techniques (domain randomization, actuator modeling, physics alignment)
- Strong analytical skills and track record of publications or conference contributions
- Fluent English communication skills
- Proficiency in German
What We Do
NEURA Robotics is a German high-tech company founded in 2019 in Metzingen near Stuttgart with the vision to revolutionize the world of robotics. More than 180 team members from over 30 countries are working on advanced technologies in the fields of environmental perception, drive and control technology, material science, mechanical design, and artificial intelligence. We are expanding the cognitive capabilities of robots and make breakthrough advances in a variety of areas to bring robots and humans closer together, making many areas of work more attractive, creative, and social again. That's why everything we do runs under the guiding principle "we serve humanity". In a very short period of time, NEURA Robotics has developed robots and technologies that are characterized above all by their outstanding performance as well as safe and human-centred way of working. In this way, a wide variety of application fields can be covered, from intelligent production to medical technology. All major robot components are developed and designed in-house. Imprint: https://www.neura-robotics.com/legal Privacy: https://www.neura-robotics.com/privacy
