RL Scaling Engineer - Long-Horizon RL at Frontier Scale

Anthropic is seeking a Research Engineer for RL Scaling Science to design and run large-scale RL experiments, advancing benchmarks, and translating validated findings into production training recipes. The role blends research rigor with engineering practicality, addressing bottlenecks and ensuring measurable long-horizon progress.

You will work at the boundary of research and systems, collaborating with adjacent RL teams to strengthen the end-to-end training stack and drive scalable, responsible