Applied Research - Agentic RL
Core
Building reinforcement learning systems for tool-using, multi-agent, and long-horizon AI agents that reason over complex tasks and improve from experience.
Role type
Applied Research Scientist (Agentic RL)
Builds
Open-source RL systems, scalable experimental harnesses, and production-grade agent training pipelines.
Domain
Frontier AI, Reinforcement Learning, Multi-Agent Systems
Deliverable
production ML models
Required skills
Reinforcement Learning, Scalable ML Systems, Agent Environments, Experimental Design, Python, Inference/Training Infrastructure
Preferred skills
Publications, Open-source contributions, Benchmarks, Shipped products
Technologies
Python, RL frameworks, Multi-agent environments
Responsibilities
Design credit-assignment methods for multi-agent trajectories, build scalable training environments for long-horizon tasks, detect and mitigate reward hacking, develop reusable evaluation harnesses, connect algorithm choices to training/inference behavior.
Seniority
Senior, hands-on IC
