Research Engineer, Pretraining Scaling - London

Overview

Research Engineer on Anthropic’s ML Performance and Scaling team, ensuring production pretrained frontier models train reliably, efficiently, and at scale.

What you'll do

• Own critical production pretraining pipeline aspects, including model operations, optimization, observability, and reliability.
• Debug and resolve complex full-stack issues spanning hardware, networking, training dynamics, and evaluation infrastructure.
• Design and run experiments to improve training efficiency, step time, uptime, and model performance.
• Respond to on-call incidents during model launches and coordinate solutions across teams.
• Build and maintain production logging, monitoring dashboards, and evaluation infrastructure.
• Add capabilities to the training codebase, such as long-context support or novel architectures.
• Document systems and debugging approaches to build institutional knowledge.

What you'll need

• Hands-on experience training large language models or deep expertise with JAX, TPU, PyTorch, or large-scale distributed systems.
• Enjoy a roughly 50/50 research-and-engineering split.
• Be excited about being on-call for production systems and working long days during launches.
• Excel at debugging complex, ambiguous problems across multiple layers of the stack.
• Communicate clearly and collaborate effectively, including across time zones and during high-stress incidents.
• Be passionate about the work and refining craft as a research engineer.
• Care about the societal impacts of AI and responsible scaling.

Details

• Role requires working in-office 5 days per week in London.
• Compensation range is £260,000 to £630,000 GBP annually.
• Annual salary range is stated; no education field minimum beyond a role-relevant field.