全栈研发工程师/RL Environments架构师-AI数据服务平台
Core
Build a modular, high-extensibility RL environment production line and core platform for Agent reinforcement learning, supporting massive training throughput, stability, and observability.
Role type
Senior IC full-stack R&D engineer / RL Environments Architect
Builds
Core infrastructure for Agent training, including simulation environments, task design, and state/reward modeling
Domain
Artificial Intelligence / Reinforcement Learning / Large Model Training
Deliverable
production ML models
Required skills
Python, Go, Agent development, Simulation environment construction, Task design, State modeling, Reward modeling, System scalability, Observability
Preferred skills
Large model training paradigms, Sim2Real migration, Cross-functional collaboration
Responsibilities
Design RL environment quality standards and governance mechanisms, Collaborate with algorithm research teams to abstract real-world business flows into reproducible simulation environments, Optimize system throughput and stability for high-volume training
Seniority
Senior, hands-on IC
