Applied Research – Robotics

This is a fully remote position, open to applicants in United States, +2 more locations.

📋 Description

• Collaborate with an external robotics team to create environments, evaluate systems, and implement post-training methodologies.

• Train and assess agents utilizing reinforcement learning, world modeling, imitation learning, planning, and multimodal perception techniques.

• Incorporate feedback obtained from simulations and deployed robotic systems.

• Examine behavior under conditions of partial observations, extended timeframes, fluctuating environments, as well as hardware and safety limitations.

• Develop simulations and real-world assessments tailored for long-horizon embodied tasks.

• Integrate demonstrations, interventions, offline data, and online feedback into comprehensive training loops.

• Educate embodied agents to recover from partial failures, adapt to changing scenes, and navigate uncertain states.

• Manage both simulation and real-world environments for training, evaluation, and sim-to-real comparisons.

• Lead long-horizon agents capable of perception, planning, action, memory, and recovery under hardware restrictions.

• Create reward systems and evaluation metrics focused on task success, safety, adaptability, and continuous improvement.

• Engage directly with customers, users, or research collaborators while upholding a meaningful research agenda.

⛳️ Requirements

• Relevant experience in robotics, embodied AI, reinforcement learning, imitation learning, control systems, world modeling, simulation, computer vision, agent evaluation, or production systems that interact with the physical environment.

• Practical experience in constructing and troubleshooting model-training, post-training, agent evaluation, or closely related experimental processes.

• Strong experimental judgment demonstrated through meaningful baselines, meticulous trace inspection, clear attribution of failures, and the adaptability to change direction based on evolving evidence.

• Proficiency in Python.

• Ability to create artifacts that can be executed, inspected, and enhanced by other researchers and engineers.

• Comfort in managing ambiguous challenges while keeping partners and team members informed.

• Proven track record of exceptional work showcased through publications, open-source projects, benchmarks, or successfully launched products.

🏝️ Benefits

• Equity incentives.

• Flexible work arrangements (San Francisco, New York, London, or remote).

• Visa sponsorship and relocation assistance.

• Budget for professional development.

• Team off-sites and attendance at conferences.