Doctoral Researcher in Humanoid Robotics, 3D Perception and Embodied AI

Develop perception and learning methods for humanoid robots in autonomous manufacturing. The doctorate combines RGB-D data, scene understanding, synthetic data, simulation-to-reality transfer and robot manipulation, with experimental validation on physical platforms. The project is hosted by the Advanced Manufacturing Lab and includes collaboration with researchers and industrial partners. Apply through the ETH online recruitment portal. The vacancy is currently advertised with an active application route; no closing date is stated.

We are looking for a highly motivated and independent researcher with strong analytical and programming skills, scientific curiosity, and a passion for intelligent robotic systems.
You should have:
A Master's degree in Robotics, Computer Science, Mechanical Engineering, Electrical Engineering, Artificial Intelligence, or a closely related field
A strong foundation in robotics, computer vision, machine learning, or related computational methods
Excellent programming skills, particularly in Python and preferably also C++
Experience with machine learning frameworks, such as PyTorch, and the development, training, or evaluation of learning-based models
A solid understanding of robotics fundamentals, including coordinate transformations, kinematics, motion planning, or robotic manipulation
Practical experience in at least one relevant area, such as 3D computer vision, robotic perception, robotics simulation, or robot learning
Experience with scientific experimentation, quantitative evaluation, and the analysis of experimental results
Familiarity with Linux-based development environments, Git, and collaborative software development practices
The ability to independently investigate challenging research questions, develop new ideas, and systematically evaluate potential solutions
A strong motivation to pursue doctoral research and contribute to scientific publications
Fluent English, both written and spoken
Knowledge of Korean and/or German is an advantage
The following qualifications would be an advantage:
Experience with RGB-D cameras, depth sensing, point clouds, 3D reconstruction, or semantic scene understanding
Familiarity with robotics frameworks such as ROS/ROS 2
Experience with robotics simulation platforms such as NVIDIA Isaac Sim, MuJoCo, Gazebo, or similar environments
Experience with synthetic data generation, dataset annotation, or simulation-to-real transfer
Knowledge of reinforcement learning, imitation learning, or learning-based robotic manipulation
Familiarity with vision-language models (VLMs), large language models (LLMs), or language-conditioned robotic policies
Experience with multi-robot systems, humanoid robotics, or collaborative manipulation
Hands-on experience integrating sensors, programming robots, or conducting experiments on physical robotic platforms
Familiarity with Docker, automated testing, CI/CD pipelines, or reproducible machine learning workflows
Previous research experience demonstrated through a Master's thesis, research projects, publications, or contributions to scientific conferences
We particularly encourage applications from candidates who combine strong theoretical foundations with practical programming skills and an interest in validating their research on real robotic systems.