Reinforcement Learning Engineer (Autonomous Driving)
Core
Develop reinforcement learning methods to improve the performance of end-to-end driving models by connecting model learning, closed-loop simulation, and real-vehicle validation.
Role type
Senior IC reinforcement learning engineer (autonomous driving)
Builds
End-to-end driving models and closed-loop simulation environments
Domain
Autonomous driving, robotics, machine learning
Deliverable
production ML models
Required skills
Reinforcement learning, policy optimization, reward design, credit assignment, simulation-based training, multi-agent environments, Python, PyTorch
Preferred skills
Self-play training, multi-agent RL, GRPO/PPO post-training, vector/occupancy-based simulators, world model rollout planning, distributed training, sim-to-real validation
Technologies
PyTorch, Python, GRPO, PPO
Responsibilities
Design VLA policy and simulator observation/action interfaces; Develop closed-loop RL post-training using policy optimization techniques; Develop self-play training structures in driving simulators; Rollout policies in future scenarios and implement action selection based on safety constraints; Validate policy safety and generalization using closed-loop benchmarks.
Seniority
Senior, hands-on IC
