PhD – Skalierbare und effiziente Reinforcement-Learning-Methoden für Physical AI (f/m/div.) (w/m/div.)
Core
Researching scalable and efficient Reinforcement Learning methods for Physical AI to enable reliable autonomous decision-making systems with reduced computational requirements.
Role type
PhD Researcher (Physical AI / Reinforcement Learning)
Builds
Next-generation efficient autonomous decision-making systems for industrial applications (robotics, automated driving).
Domain
Artificial Intelligence, Robotics, Physical AI
Deliverable
research
Required skills
Reinforcement Learning, Imitation Learning, World Models, Model Compression, Python, PyTorch, JAX, TensorFlow, C, ROS 2
Preferred skills
Safe RL, Offline RL, Multi-Agent RL, MLOps, CI/CD, Software development in large teams, Publications at top ML/Robotics conferences
Technologies
PyTorch, Hugging Face, JAX, TensorFlow, ROS 2
Responsibilities
Develop AI agents that continuously improve and adapt to new situations; bridge the gap between simulation and reality; optimize resource efficiency for real-world deployment.
Seniority
PhD Candidate (Researcher)
