Summary

Research engineer on anthropic’s rl scaling science team, studying how reinforcement learning behaves as models scale and turning that understanding into training recipes. You will design and run large-scale rl experiments, build long-horizon benchmarks, and debug issues at the research/infrastructure boundary, with a hybrid office policy.

Responsibilities

• Design, run, and interpret large-scale rl experiments
• Investigate rl improvements as horizon and compute grow
• Build and maintain long-horizon rl benchmarks
• Translate validated findings into production training recipes
• Debug failures at the research and infrastructure seam
• Partner with adjacent rl teams to advance rl stack