Bear Robotics

• →None

• Develop, train and deploy robot policies for contact-rich manipulation, bimanual tasks, locomotion-aware manipulation and other whole-body behaviours.
• Combine and advance techniques across imitation learning, reinforcement learning, visuomotor learning, foundation or vision-language-action models, planning and classical control where each is most effective.

• Take learning systems end to end: problem definition, data collection, representation and model choice, training, simulation, evaluation, on-robot testing and production deployment.

• Build scalable data pipelines for teleoperation, demonstrations, autonomous rollouts, labelling, replay and failure analysis; improve the data flywheel as deployments grow.

• Create rigorous offline and on-robot evaluation, including task-level success metrics, robustness testing, regression suites and clear go/no-go criteria for pilots.

• Close the sim-to-real and lab-to-site gaps by improving domain randomisation, system identification, calibration, perception robustness, adaptation and recovery behaviour.
• Optimise models and runtime systems for real robot constraints, including latency, compute, reliability, observability and safe fallback behaviour.

• Diagnose failures systematically across learning, perception, controls, hardware and data rather than treating model performance in isolation.

• Set technical direction for substantial projects, write clear design documents and raise the engineering quality of the team through reviews, mentoring and practical standards.
• Collaborate closely with controls, perception, simulation, hardware, product and field operations teams, including hands-on work with robots in the lab and at pilot sites.

• Performs other duties or takes on specialized responsibilities as assigned.

• Strong experience developing learning-based systems for physical robots, with evidence of getting methods to work reliably on hardware rather than only in simulation or benchmark environments.
• Deep practical knowledge in one or more relevant areas: robot learning, imitation learning, reinforcement learning, visuomotor policy learning, manipulation, whole-body control or motion planning.
• Excellent Python and modern machine-learning engineering skills, including substantial experience with PyTorch or an equivalent framework.
• Experience working with robotics software and systems such as ROS/ROS 2, real-time or near-real-time pipelines, robot sensors, calibration and hardware debugging.
• A strong grasp of robotics fundamentals, including kinematics, dynamics, state estimation, control and the physical realities of contact and actuation.
• Ability to design good experiments, select meaningful metrics, interpret failures and make sound trade-offs between research ambition and deployment reliability.
• Track record of independently leading technically ambiguous work and collaborating effectively across disciplines.
• Clear written and verbal communication, with the judgement to explain complex technical decisions to both specialists and non-specialists.

• Humanoid robots, dual-arm manipulation, dexterous manipulation or contact-rich tasks.
• Large-scale robot datasets, teleoperation systems, behavioural cloning, diffusion policies, transformers or vision-language-action models.
• Reinforcement learning for robotics, including offline RL, model-based methods, residual learning or safe exploration.
• Simulation platforms and sim-to-real workflows, such as MuJoCo, Isaac Sim/Lab, Drake or equivalent tools.
• Deployment of GPU-accelerated inference on embedded or edge hardware, model optimisation and performance profiling.
• Safety-aware learning, uncertainty estimation, out-of-distribution detection, recovery policies or runtime monitoring.
• Shipping robotics capabilities into customer or pilot environments and learning from field data.

The physical demands described here are representative of those that must be met by an employee to successfully perform the essential functions of this job. Reasonable accommodations may be made to enable individuals with disabilities to perform the essential functions.

• Prolonged periods of sitting/standing at a desk and working on a computer. The employee routinely is required to sit; stand, walk; talk and hear; use hands to keyboard.
• Specific vision abilities required by this job include close vision, color vision, peripheral vision, depth perception, and ability to adjust focus.
• Ability to lift 30 lbs.
• Occasional travel to partner or pilot sites may be required.

• An MSc or PhD in robotics, machine learning, computer science, engineering or a related field, or equivalent practical experience.