DiDi's autonomous driving unit is looking to develop Level 4 autonomous driving technology to make transportation safer and more efficient by integrating AD technology into a shared-mobility fleet.
Requirements
- Expert proficiency in C++/Python
- Deep expertise in motion planning decision-making (behavior planning, world modeling, agent reaction & reasoning under uncertainty)
- Background in POMDPs, reinforcement learning, or game theory is a bonus
- Experience with simulation frameworks for policy and behavior validation
- 5+ years of experience in autonomous systems, focusing on decision-making, agent & world modeling and planning
Responsibilities
- Architect the geometric reasoning system for long-horizon planning and multi-agent interactions, resolving dense urban stuck situation and make vehicle drive with both efficiency and safety.
- Lead development of decision-making algorithms for ambiguous scenarios.
- Drive the integration of prediction and planning to enable proactive behaviors.
- Establish metrics and validation frameworks for decision quality and safety.
- Provide technical leadership and mentor junior & senior engineers within the team.
Other
- M.S. or Ph.D. in CS, Robotics, or related field
- 5+ years of experience in autonomous systems
- Experience in leading a decision-making team in a commercial AV program
- Publications in robotics/AI conferences on decision-making or planning
