- 搭建并维护用于端到端和 VLA 自动驾驶模型的强化学习闭环训练流程。

- 设计和实现支持 RL 闭环训练与评测的仿真环境。

- 开发高效可扩展的工具链，包括数据管理、实验调度和性能监控。

- 对强化学习算法进行优化，提升训练效率、可扩展性及实时部署能力。

- 与研究团队协作，将新的 RL 方法集成到闭环系统中。

- 记录开发流程与基准结果，提供部署相关的技术支持。

- Build and maintain closed-loop reinforcement learning training pipelines for E2E and VLA autonomous driving models.

- Design and implement simulation environments to support RL-based closed-loop training and evaluation.

- Develop scalable toolchains for dataset management, experiment orchestration, and performance monitoring.

- Optimize RL algorithms for efficiency, scalability, and real-time deployment.

- Collaborate with research teams to integrate new RL methods into the closed-loop system.

- Document development workflows, benchmark results, and provide technical support for deployment.

1.计算机、机器学习、自动化、机器人等相关专业硕士或博士学历。

2. 具备强化学习、仿真环境、大规模训练流程等相关经验。

3. 熟悉自动驾驶仿真平台（如 CARLA、LGSVL、SUMO, GPUDrive, Waymax）或机器人仿真环境。

4. 具备扎实的软件工程能力，精通 Python/C++，有分布式训练与工具链开发经验。

5. 熟悉容器化技术（Docker、Kubernetes）及实验管理工具。

6. 具备良好的问题解决能力和团队协作精神，自驱动。

7. 具备良好的英文读写能力。

1. Master’s/Ph.D. degree in Computer Science, Software Engineering, or related fields.

2. Solid background in reinforcement learning, simulation environments, and large-scale training pipelines.

3. Hands-on experience with autonomous driving simulators (e.g., CARLA, LGSVL, SUMO, GPUDrive, WayMax) or robotics simulators.

4. Strong software engineering skills in Python/C++; experience in distributed training and toolchain development.

5. Familiarity with containerization (Docker, Kubernetes) and experiment management tools.

6. Good problem-solving skills, self-driven, and team-oriented.

7. English reading/writing proficiency.