Doctoral Positions in Robot Learning and Soft, Musculoskeletal, and Biohybrid Robotics

The Soft Robotics Lab within the Institute of Robotics and Intelligent Systems at ETH Zurich is inviting applications for several doctoral positions. Our lab's goal is to build, model, and control robots in a fundamentally different way, so that they become more flexible, dexterous, capable, and adapt better to their environment. We work along four directions: soft and musculoskeletal robotics, biohybrid living systems, dexterous manipulation and robot learning, and simulation for embodied AI. We are looking for exceptional candidates in any of them. This round we especially want two profiles: people who design and build the robots, and people who make policies run on them.

We do not hire against a narrow project description. Your thesis topic is something we shape together in your first months. Tell us which of our directions pulls at you, and what you would want to build.

Project background

Today's robots are mostly rigid, fragile, and a world apart from the agility and resilience of biological bodies. Our bet is that the next generation of robots will be soft, musculoskeletal, and in part alive. They will be built to make contact with the real world rather than to avoid it. We pursue this across four directions, and a strong candidate will find a home in one of them and borrow from the others.

Soft and musculoskeletal robotics. We build bodies from compliant structures, bones, joints, and tendon-like actuation. Our electrohydraulic musculoskeletal leg jumps, moves fast, and adapts to terrain at roughly 1.2% of the energy a motor-driven leg needs (Nature Communications, 2024). Our low-voltage HASEL actuators run near 1100 V, are safe to touch, and work untethered and underwater (Science Advances, 2024). We recently extended these muscles to full antagonistic motion ranges (ICRA 2025) and to a sensorless, inherently compliant anthropomorphic hand driven entirely by electrohydraulic actuation (IROS 2026).

Biohybrid living systems. We grow engineered muscle and use it to actuate machines. We bioprinted multicellular muscle-tendon units that transmit force along a real musculoskeletal path (Science Advances, 2025), embedded sensors directly into muscle for closed-loop control of proprioceptive biohybrid robots (Advanced Intelligent Systems, 2025), and established functional volumetric bioprinting with xolography (Advanced Materials, 2026). Co-optimized volumetric muscle designs for large dynamic deformations are in press at Nature Communications (Balciunaite et al., 2026). The same fabrication line reaches clinical work: with University Hospital Zurich we printed implantable reinforced cardiac tissue patches (Advanced Materials, 2025).

Dexterous manipulation and robot learning. We build hands and the policies that run them. One of our initial hand designs is now commercialized through our spin-off Mimic Robotics. ORCA is our open-source, reliable, and cost-effective anthropomorphic hand for uninterrupted dexterous task learning (IROS 2025). On that hardware we work on imitation learning and diffusion policies, cross-embodiment skill transfer through latent action diffusion (ICRA 2026), sample-efficient reinforcement learning and policy fine-tuning directly on the real robot, vision-language-action models for contact-rich tasks, and tactile representation learning on our high-resolution sensorized skin (ICRA 2024). We also build controllable dexterous world models for training and evaluation, and a benchmark of dexterity for anthropomorphic hands. Whichever side you come from, the offer is the same: the hand, the skin, the simulator, and the people who designed all three sit in one room. Build a mechanism here and someone will have a policy running on it within weeks. Build a policy here and you can change the mechanism when the mechanism is what is wrong.

Simulation, fabrication, and embodied AI. Building these robots requires tools that did not exist. Vision-Controlled Jetting prints rigid skeletons, soft tissue, tendons, and sensors in one pass, including a full musculoskeletal hand and forearm (Nature, 2023). We close the sim-to-real gap with learned residual physics (RA-L, 2024, Best Paper Award), and we released SORS, a modular high-fidelity soft-robot simulator, at RoboSoft 2026.

Underwater and aerial systems run through all of this, from SoFi and tendon-driven swimmer digital twins to our open-source soft aerial manipulation platform (CoRL 2024).

Job description

Depending on your direction, your work will emphasize different parts of the following. All of it happens in a lab where hardware, biology, and learning sit in the same room.

• You will take a research idea from concept to a working system: designing the architecture, building it, integrating sensing and control, and validating it through systematic real-world experiments
• If your focus is systems design, you will design and build the robots themselves: mechanisms and compliant structures, actuators and tendon routing, embedded electronics and motor control, and the sensor integration that makes a machine measurable. You will own a system from CAD through fabrication to hardware that still runs six months later
• If your focus is learning, you will develop policies and perception that run on real, compliant, contact-rich hardware: imitation learning, real-world reinforcement learning and fine-tuning, sim-to-real transfer, and multimodal representations that use touch as well as vision. You will help define the benchmarks that make such claims measurable
• Each design cycle feeds the next, so rapid prototyping, measurement, and iteration sit at the heart of every project
• Drawing inspiration from biological musculoskeletal systems, you will engineer how bones, joints, tendons, and muscles can be recreated with compliant materials, artificial actuators, or living tissue, and how their interplay produces strength, dexterity, and robustness
• You will build robots that derive much of their capability from their embodiment, achieving rich, adaptive behavior with less reliance on complex centralized control
• If your focus is biohybrid systems, you will work in our biological laboratories at ETH, culturing and bioprinting tissue and turning it into a controllable actuator
• You will publish at the top venues in the field, release open-source hardware and code where it helps the community, and present your work internationally
• You will supervise semester and Master's projects, which is how most of our doctoral students learn to lead work, and you will collaborate day to day across our hardware, muscles, and machine learning teams
• You will complete a doctorate at ETH Zurich alongside the research, including the coursework of the D-MAVT doctoral programme and a share of teaching in one of our courses

Profile

You are curious, highly motivated, and independent, and you want to make a real difference with your research
You work best in a team. You are respectful, and you thrive when you can collaborate with others and support them

Through your prior experience, you have ideally already demonstrated:

• A completed or nearly completed Master's degree in mechanical engineering, robotics, mechatronics, electrical engineering, computer science, materials science, bioengineering, or a closely related field
• Depth in at least one of: hands-on system building and mechatronics (CAD, FEA, 3D printing, machining, molding, electronics, motor control, robot integration), machine learning and control for real robots, soft or bio-inspired materials and actuators, tissue engineering and biofabrication, or physics-based simulation
• Evidence that you carry work through to reality. In a thesis, a semester project, a competition team, or an internship, you have taken something from first idea to a result that worked on real hardware, not only to a simulation benchmark
• Substantial project or thesis work we can read, and ideally a publication or preprint, though we do not expect one at this stage
• Strong written and spoken English, and comfort explaining your work clearly
• Strong motivation for interdisciplinary, experimental research and the curiosity to develop new skills throughout the doctorate
• A collaborative, supportive mindset and the drive to make a real difference with your research
• A clear idea of what you would want to work on with us, and why it matters

Nobody arrives with all of this. Show us the few things you are already good at and the appetite to learn the rest.

Workplace

We offer

• Supervision that is actually available: weekly one-on-one meetings, a lab that sits in one space, and co-supervision from senior doctoral students and postdocs in your direction
• A conference travel budget, so you present your own work at ICRA, IROS, CoRL, RSS, RoboSoft, or the venue that fits your topic, and you go more than once
• A world-class lab with state-of-the-art fabrication and test facilities: in-house multi-material 3D printing, a machine shop, an electronics lab, our own biological laboratories, over 20 dexterous robotic hands and 12 robot arms for real-world data collection and teleoperation, high-resolution tactile skins, and a dedicated GPU cluster with eight NVIDIA H200 GPUs alongside access to ETH's central compute and the Alps supercomputer at CSCS
• A multidisciplinary team of mechanical engineers, materials scientists, biologists, and machine learning researchers, embedded in the broader robotics ecosystem at ETH Zurich
• Strong ties to industry and to our spin-offs, and support for turning your research into something that leaves the lab. Our doctoral graduates have gone to faculty positions, to top industry labs, and into founding companies
• Your job with impact: become part of ETH Zurich, which not only supports your professional development, but also actively contributes to