Research Engineer, Pretraining Scaling - London

About Anthropic

• Creating reliable, interpretable, and steerable AI systems
• Building beneficial AI systems
• Team of committed researchers, engineers, policy experts, and business leaders

About the Role:

• Training production pretrained models
• Ensuring frontier models train reliably, efficiently, and at scale
• Working at the boundary between research and engineering
• Performance optimization, hardware debugging, experimental design, and launch coordination

Responsibilities:

• Owning critical aspects of the production pretraining pipeline
• Debugging and resolving complex issues across the full stack
• Designing and running experiments to improve training efficiency
• Responding to on-call incidents during model launches
• Building and maintaining production logging and monitoring infrastructure
• Collaborating with teammates across locations
• Contributing to institutional knowledge through documentation

You May Be a Good Fit If You:

• Have hands-on experience training large language models
• Enjoy both research and engineering work
• Thrive under pressure and enjoy solving hard problems
• Excel at debugging complex, ambiguous problems
• Communicate clearly and collaborate effectively
• Are passionate about refining your craft as a research engineer
• Care about the societal impacts of AI

Strong Candidates May Also Have:

• Previous experience with large language models or ML frameworks at scale
• Contributions to open-source LLM frameworks
• Published research on model training or ML systems
• Experience with production ML systems or observability tools
• Background in roles requiring technical depth and operational excellence

What Makes This Role Unique:

• Highly operational with hands-on experience in large training runs
• Intense learning opportunities with world-class researchers and engineers
• Operational intensity offers unique rewards for those who thrive in this environment

Perks and Benefits:

• Competitive compensation and benefits
• Optional equity donation matching
• Generous vacation and parental leave
• Flexible working hours
• Lovely office space for collaboration