Reinforcement Learning Engineer: Build Scalable Safe AI

Anthropic; Greater London; Hybrid; GBP 260,000 - 630,000; Full time

Benefits offered by this job

Competitive compensation

Equity donation matching

Generous vacation

Parental leave

Flexible working hours

Job summary

Anthropic is hiring a Research Engineer in Reinforcement Learning to advance capabilities and safety of large language models. You will blend research and engineering, implement novel approaches, and contribute to research direction, focusing on agentic models, tool use, and reasoning improvements for open-ended tasks.

You will architect RL infrastructure, develop evaluation environments, and collaborate across teams to push state-of-the-art while ensuring safety and scalability across GPU

Qualifications

• Bachelor’s degree or equivalent in a relevant field.
• Experience in machine learning research is preferred.
• Strong coding and systems design skills are essential.
• Ability to balance research exploration with engineering implementation.

Responsibilities

• Architect and optimize reinforcement learning infrastructure and experiment management.
• Design, implement, and test training environments and evaluation methodologies for RL agents.
• Drive performance improvements via profiling, optimization, and benchmarking across the stack.
• Collaborate across research and engineering teams to build scalable AI research infrastructure.

Skills

Python

Async programming

Trio

PyTorch

TensorFlow

JAX

ML research

Pair programming

Code quality

Systems design

Education

Bachelor's degree

Tools

Kubernetes

Rust

C++

Job description

Anthropic is hiring a Research Engineer in Reinforcement Learning to advance capabilities and safety of large language models. You will blend research and engineering, implement novel approaches, and contribute to research direction, focusing on agentic models, tool use, and reasoning improvements for open-ended tasks.

You will architect RL infrastructure, develop evaluation environments, and collaborate across teams to push state-of-the-art while ensuring safety and scalability across GPU