RL Scaling Research Engineer

Anthropic; City Of London; Hybrid; GBP 375,000 - 640,000; Full time

Job summary

Anthropic is seeking a Research Engineer to join the RL Scaling Science team. You will design and run large-scale experiments, build benchmarks, and translate findings into production training recipes.

The role sits at the boundary between research and engineering, with open problems and frontier-scale experiments, offering a path from robust results to production. Strong collaboration with adjacent RL teams across research and infrastructure is essential.

Qualifications

• Strong empirical RL research and large-scale ML training.
• Ability to run end-to-end experiments from design to interpretation.
• Proficiency in Python and experience with large-scale ML systems.
• Comfort at the research/systems interface and debugging at scale.
• Commitment to responsible AI and societal impacts.

Responsibilities

• Design, run, and interpret large-scale RL experiments.
• Study how RL scales with horizon, compute, and model size.
• Build benchmarks for long-horizon RL that are measurable and reproducible.
• Translate validated findings into production training recipes.
• Collaborate with RL teams across research and engineering.
• Debug issues at the seam between research and infrastructure.

Skills

RL research

Experiment ownership

Python

Debugging at scale

AI ethics

Tools

Python

Distributed ML systems

Job description

Anthropic is seeking a Research Engineer to join the RL Scaling Science team. You will design and run large-scale experiments, build benchmarks, and translate findings into production training recipes.

The role sits at the boundary between research and engineering, with open problems and frontier-scale experiments, offering a path from robust results to production. Strong collaboration with adjacent RL teams across research and infrastructure is essential.