Requirements

• Hold an MSc or PhD in Robotics, Machine Learning, Control Engineering, or a closely related field, focusing on reinforcement learning and robot control.
• Hands-on experience training and deploying RL-based controllers on real robotic hardware.
• Deep familiarity with modern RL techniques for continuous control, including model-free methods (PPO, SAC, TD3), model-based RL, hierarchical RL, sim-to-real transfer, domain randomisation, and curriculum learning.
• Solid understanding of robot dynamics, kinematics, and classical control theory (e.g. PID, model predictive control, trajectory optimisation), and how to integrate them with learned approaches.
• Proficiency in building and using physics-based simulators such as MuJoCo, Isaac Gym/Isaac Lab, PyBullet, or Gazebo.
• Strong software engineering skills with clean and well-structured code in Python and/or Rust or modern C++, including experience with AI software deployment, testing, QA, and monitoring.
• Excellent communication skills for effectively reporting and presenting research findings internally and externally.
• Passion for staying updated with current research and re-implementing or extending state-of-the-art papers.

Nice to Haves

• PhD in Robotics, Reinforcement Learning, Control Engineering, or related fields with publications in top-tier venues such as CoRL, ICRA, NeurIPS, ICLR, IROS, or RSS.
• Experience developing controllers for highly dynamic robotic systems in complex contact and environmental conditions.
• Experience with vision-based perception for robotics control, including depth estimation, visual odometry, or visuomotor policy learning.
• Familiarity with low-level motor control interfaces and real-time embedded system constraints.
• Experience with online adaptation and meta-learning techniques for robotic adaptation to changing environments.
• Experience with sensor fusion (IMU, proprioception, exteroception, vision) to enhance learned control policies.

What You'll Be Doing

• Shaping AI capabilities for autonomous robotics in challenging sectors with hardware constraints and real-world robustness needs.
• Designing, training, and deploying reinforcement learning-based controllers for robots operating in complex, unstructured, and dynamic environments where classical control fails.
• Enabling robots to perceive and understand surroundings by fusing rich sensory inputs, including vision, for robust and adaptive control.
• Owning the entire pipeline from simulation to deployment, ensuring learned policies perform robustly and efficiently under tight hardware constraints.

Perks and Benefits

• Competitive salary and VSOP options.
• Relocation support up to €2,500 and 4 weeks of temporary accommodation.
• Annual learning allowance of €500/£450.
• Health and wellness benefits including gym membership and mental health support (Nilo.health).
• Regular company events and monthly social allowances.
• Enhanced parental leave: 22 weeks fully paid for primary caregivers and 6 weeks for secondary caregivers.
• Family support including 5 days of paid family emergency leave, 100% remote work option during pregnancy, and phased return to work.