Research Engineer / Research Scientist, RL Frontiers

The role

You will develop next-generation architectures and RL algorithms and scale them to frontier-scale runs. You will build experimental infrastructure and own the end-to-end performance of large RL runs.

What you will do

• Study how RL training and sampling scale with model size, context length, and compute
• Develop next-generation model architectures and RL algorithms
• Take promising small-scale results to frontier-scale runs
• Build experimental infrastructure for fast, reproducible comparisons
• Own end-to-end performance of largest RL runs from research code to hardware
• Build performance and cost models for proposed architecture and algorithm changes
• Investigate training dynamics at scale including instabilities and throughput regressions

Required qualifications

• Deep familiarity with modern transformer language models
• Hands-on experience training large models in a distributed setting
• Track record of original technical work in ML training or systems
• Ability to design rigorous experiments at scale
• Ability to reason quantitatively about compute, memory, and communication costs
• Strong programming skills in Python and JAX or PyTorch
• Comfort reading and changing code at every layer of the stack
• Bachelor's degree or an equivalent combination of education, training, and/or experience
• Field of study relevant to the role as demonstrated through coursework, training, or professional experience

Nice to have

• Research experience in reinforcement learning, optimization, or large-scale training
• Experience developing RL algorithms for language models
• Experience with scaling laws or other quantitative models of training efficiency
• Experience designing or modifying transformer architectures beyond standard configurations
• Experience scaling training to large fleets of accelerators and debugging problems at scale
• Deep understanding of numerics in large-scale training
• Familiarity with how GPU or TPU performance characteristics shape architecture and algorithm choices
• Experience with C++ or Rust

Benefits

• Competitive compensation and benefits
• Optional equity donation matching
• Generous vacation and parental leave
• Flexible working hours
• Office space

Before you apply

• We do sponsor visas, but not for every role and candidate
• Recruiters only contact you from @anthropic.com email addresses

About the company

Anthropic is a public benefit corporation headquartered in San Francisco. Its mission is to create reliable, interpretable, and steerable AI systems.