大模型Multi-Agent RL算法研究员 - Seed Model
Core
Researching and optimizing long-horizon tasks using Large-Scale RL and Multi-Agent RL to advance general intelligence and support applications like Doubao and Jiemeng.
Role type
Senior Research Scientist (Multi-Agent RL)
Builds
Next-generation Multi-Agent frameworks and training algorithms for internal use and enterprise clients.
Domain
Artificial Intelligence / Reinforcement Learning / Multi-Agent Systems
Deliverable
production ML models
Required skills
Reinforcement Learning, Multi-Agent Systems, Large-Scale RL, ML Systems (MLsys), Agent Frameworks, Experimental Design
Preferred skills
Self-Evolving Algorithms, Multi-Agent Harness Design, Open-source framework expertise (Verl, Slime, Megatron)
Technologies
Verl, Slime, Megatron, Claude Code, Codex
Responsibilities
Optimize long-horizon tasks using Large-Scale and Multi-Agent RL; Innovate Multi-Agent training techniques and internal frameworks.
Seniority
Senior, hands-on IC
