About the company
Anthropic's mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole.
About the role
Reinforcement learning is how Claude learns to reason, write code, and act autonomously over long horizons. The RL Scaling team works on how RL scales: what happens to throughput, stability, and learning efficiency as models get larger, episodes get longer, and compute grows by orders of magnitude.
Responsibilities
- Study how RL training and sampling scale with model size, context length, and compute
- Develop next-generation model architectures and RL algorithms, and make them run efficiently at frontier scale
- Take promising small-scale results to frontier-scale runs, and diagnose why they behave differently
- Build the experimental infrastructure that sets research velocity
- Own end-to-end performance of our largest RL runs, from research code down to the hardware
- Build performance and cost models for proposed architecture and algorithm changes
- Investigate training dynamics at scale, including instabilities, divergence, and throughput regressions
Requirements
- Deep familiarity with modern transformer language models, including their architecture, training dynamics, and large-scale optimization
- Hands-on experience training large models in a distributed setting, including tradeoffs between data, tensor, and pipeline parallelism
- A track record of original technical work in ML training or systems
- Ability to design rigorous experiments at scale, including baselines, ablations, and statistical care
- Ability to reason quantitatively about the compute, memory, and communication costs of a model or algorithm
- Strong programming skills in Python and JAX or PyTorch
Preferred qualifications
- 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
- Deep understanding of numerics in large-scale training, including low-precision formats
- Familiarity with how GPU or TPU performance characteristics shape architecture and algorithm choices
- Experience with C++ or Rust
Conditions
- Annual Salary: $500,000 - $850,000 USD
- Location-based hybrid policy: all staff expected in one of our offices at least 25% of the time
- Visa sponsorship available
- Minimum education: Bachelor's degree or equivalent