Robotics Engineer
Saxon GlobalAbout the Role
We are looking for an experienced applied robotics/robot learning engineer with strong reinforcement learning expertise and hands-on experience deploying learning-based policies for complex manufacturing tasks.
The ideal candidate combines strong RL fundamentals with practical robotics engineering experience and some exposure to bimanual manipulation.
Approximately 80% of the role will focus on applied engineering, policy implementation, robot experiments, evaluation, and deployment, with 20% dedicated to research and technical exploration. The emphasis is on translating RL methods into measurable improvements on physical robots rather than pursuing research in isolation.
Key Responsibilities
· Develop, fine-tune, and evaluate VLA models and learning-based policies on physical robotic systems.
· Apply reinforcement learning to improve manipulation success rates, execution efficiency, robustness, and recovery from failures.
· Implement and evaluate appropriate approaches such as PPO, SAC, residual RL, policy adaptation, or learned recovery policies.
· Design reward functions and success detectors, including learned or VLM-based reward models, for real-world manipulation tasks.
· Apply sample-efficient methods for real-world training, such as RL from demonstrations, offline-to-online RL, and human-in-the-loop corrections.
· Set up safe real-robot RL, including action limits, workspace constraints, automated resets, and supervised exploration.
· Diagnose sim-to-real gaps and improve real-world performance using techniques such as domain randomization, system identification, and real-world fine-tuning.
· Apply learned policies to manipulation tasks, including tasks involving bimanual setups and coordinated robot execution.
· Design evaluations, analyze failure modes, and measure improvements in success rate, cycle time, generalization, and repeatability.
· Work directly with robotic hardware to run experiments, debug policies, and validate improvements.
· Supervise data collection, identify data-quality issues, and coordinate corrections to demonstrations or datasets.
· Collaborate with the simulation team and contribute to benchmarking, model evaluation, and other hands-on engineering tasks as needed.