Research Engineer Pretraining Scaling
AnthropicVisit Anthropic website
AI safety and research company building reliable, interpretable, and steerable AI systems, including the Claude product family and developer platform.
Series F+Recently funded59 current maintainers55 active leads23 new active leads26 lead step-downs8 early lead departures
San Francisco, United States
Funding history
About Anthropic
Anthropic PBC develops frontier AI systems and deploys them through Claude products and the Claude Platform, with a stated focus on safety, interpretability, and steerability.
Skills
Candidate Availability
About the Role
You will own and improve production pretraining systems, including performance, reliability, observability, and model operations. You will resolve hardware, networking, training, and evaluation issues; run efficiency experiments; respond to launch incidents; maintain monitoring systems; extend training code; and document operational knowledge.
Requirements
- Large language model training
- JAX
- TPU
- PyTorch
- Distributed systems
- Performance optimization
- Hardware debugging
- Experimental design
- Observability
- Production systems
- On-call support
- AI safety
Responsibilities
- Own production pretraining pipeline operations, performance optimization, observability, and reliability
- Debug and resolve hardware, networking, training, and evaluation issues
- Design and run experiments to improve training efficiency and model performance
- Respond to on-call incidents during model launches
- Build and maintain production logging, monitoring dashboards, and evaluation infrastructure
- Add capabilities to the training codebase
- Document systems, debugging approaches, and lessons learned
Benefits
- Optional equity donation matching
- Generous vacation
- Parental leave
- Flexible working hours
