RL Systems Engineer — Distributed & Scalable

EngineersOfAI; New York (NY); On-site; USD 180,000 - 240,000; Full time

Job summary

Anthropic is seeking a Research Engineer to join the Distributed Systems team in RL Engineering. You will work on the large-scale RL system that runs training, sampling, and environment execution, moving between layers from scheduling to data movement and networking.

Strong problem solving, first-principles thinking, and cross-team collaboration are essential. Responsibilities include designing and operating distributed RL infrastructure, improving reliability, and building automation to enable

Qualifications

• Proficient Python
• Experience with systems languages (Rust/C++/Go)
• Strong knowledge of distributed systems in production
• Ability to reason about throughput, latency, and resources
• Experience debugging complex failures across many hosts/services
• Clear technical writing and incident documentation

Responsibilities

• Design, build, and operate distributed RL systems at scale (training, sampling, environments)
• Identify bottlenecks and remove system limits across components
• Incorporate fault tolerance, isolation, and recovery in all layers
• Design resource management and autoscaling to match demand
• Develop observability to diagnose runs and performance issues
• Create automation and interfaces for engineers and automated tools
• Collaborate with researchers and performance engineers to preserve training correctness
• Document design decisions and improvements with clear writeups

Skills

Python

Rust

C++

Go

Distributed systems

Performance analysis

Debugging at scale

Technical writing

Job description

Anthropic is seeking a Research Engineer to join the Distributed Systems team in RL Engineering. You will work on the large-scale RL system that runs training, sampling, and environment execution, moving between layers from scheduling to data movement and networking.

Strong problem solving, first-principles thinking, and cross-team collaboration are essential. Responsibilities include designing and operating distributed RL infrastructure, improving reliability, and building automation to enable