5 months ago
Zürich, SwitzerlandSenior
Responsibilities
- Lead the design, training, and deployment of reinforcement learning policies for robot motion from simulation through reliable real-world operation.
- Provide technical guidance on reinforcement learning and learning-based control, mentor engineers, and establish policy development best practices.
- Own and evolve RL training infrastructure and the sim-to-real pipeline for reproducible, scalable, and rapid experimentation.
- Shape internal ML tooling and experiment management, including training dashboards and automated evaluation pipelines.
- Collaborate with stakeholders to expand the robot’s autonomous operational envelope.
- Triage field locomotion issues, identify failure patterns, and improve policy robustness using real deployment data.
- Write, deploy, and maintain Python and C++ software for the learning and locomotion stack.
Requirements
- PhD in robotics, machine learning, computer science, or a related field focused on reinforcement learning, or an equivalent track record of RL research and robotics deployment.
- Alternatively, a master’s degree in a relevant technical field and 5+ years of professional experience.
- Proven experience shipping ML models to the field and maintaining them over time.
- Strong grounding in robot control and autonomous systems, including motion control, state estimation, path planning, and actuation.
- Experience with robotic simulation tools such as Gazebo or Isaac Sim.
- Strong understanding of sim-to-real transfer, domain randomisation, reward shaping, and policy robustness techniques.
- Proficiency in Python and modern ML frameworks such as PyTorch, with working knowledge of C++.
- Strong knowledge of Linux systems and middleware frameworks for integrating learned components into larger software stacks.
- Excellent English communication skills.
- Preferred qualifications include experience training RL policies on physical robots, scalable robot motion-control architectures, navigation and autonomous mobile robot operation, agentic engineering toolchains, software architecture leadership, and physical systems such as multibody dynamics, electromechanical drive physics, energy optimization, and contact physics.
Benefits
- Fast-growing, dynamic work environment with an ambitious robotics team.
- Opportunity to contribute ideas and work on mobile robotics for industrial inspection and maintenance.
- Fair market salary and an attractive employee stock ownership plan.
- Position based in Zurich.
Categories
ML EngineeringRobotics
About ANYbotics
ANYbotics builds autonomous legged robots and software for industrial inspection and data collection in hazardous or hard-to-reach facilities. Its flagship ANYmal quadruped, including explosion-proof variants, supports routine inspections, remote operations, and predictive maintenance in energy, power, mining, and chemical plants. Founded in 2016 and headquartered in Zurich, the privately held company sells robots, software, and deployment services to large industrial operators.