Senior/Staff Machine Learning Engineer (Reinforcement Learning), Motion Planning

Responsibilities

Design, develop, and deploy learned behavior planning models using approaches such as reinforcement learning, behavior cloning, and imitation learning
Own end-to-end Reinforcement Learning pipeline from environment construction, training, and modeling.
Contribute to Reinforcement Learning Infrastructure to facilitate large-scale distributed training
Track model training, analyze model performance, identify failure modes to improve generalization across driving scenarios
Collaborate with simulation and planning teams to deploy learned behavior planning models to driving pipelines

Required skills

MS or PhD in Computer/Software Engineering, Robotics, Machine Learning or related field.
Experience developing production code in C/C++ and Python
Experience with machine learning frameworks such as PyTorch
Experience with reinforcement learning
Strong oral and written communication skills.

Preferred skills

Experience with high-throughput batched driving simulators such as GPUDrive and PufferDrive
Experience with self-play reinforcement learning
Experience with sim-to-real in the context of robotics / autonomous driving
Prior experience in learned behavior planning / prediction for autonomous vehicles

Additional Information

Compensation

$130,000 - $220,000 a year