Latest AI JobsBrowse Jobs
Anthropic logo

Research Engineer, Pretraining Scaling - London

AnthropicLondon, UKAI Research & Engineering
Apply on company site
Salary
Not listed
Job type
Full-time
Posted
30 Sept 2025
Status
Open · checked 9d ago

About this role

The Research Engineer, Pretraining Scaling position at Anthropic in London is an onsite role focused on training reliable, efficient frontier models at scale for candidates with deep machine learning systems expertise. This position suits professionals who enjoy a balanced mix of engineering and research while driving the development of safe and beneficial artificial intelligence.

Role Overview

Anthropic is hiring a Research Engineer for the ML Performance and Scaling team in London to train production pretrained models. This position sits at the intersection of research and engineering, requiring hands-on work across the entire production training stack. Team members handle performance optimization, hardware debugging, experimental design, and launch coordination. During active launches, engineers work closely together to respond immediately to critical production issues.

Key Responsibilities

Take ownership of critical production pretraining pipeline components, including model operations, reliability, observability, and performance optimization.
Troubleshoot and resolve complex full-stack issues spanning networking, hardware errors, evaluation infrastructure, and training dynamics.
Create and execute experiments aimed at increasing uptime, enhancing model performance, reducing step time, and boosting training efficiency.
Manage on-call duties during model launches by swiftly diagnosing problems and coordinating cross-team solutions.
Develop and maintain production logging, evaluation infrastructure, and monitoring dashboards.
Integrate new capabilities into the training codebase, such as novel architectures or long context support.
Cooperate with colleagues across London and San Francisco, as well as with Systems, Architectures, and Tokens teams.
Document systems, lessons learned, and debugging methodologies to build institutional knowledge.

Requirements

Background featuring hands-on experience training large language models or deep proficiency with JAX, PyTorch, TPU, or large-scale distributed systems.
Passion for both engineering and research work, with an ideal split weighted evenly between both disciplines.
Readiness to handle on-call responsibilities for production systems, work extended hours during launches, and resolve difficult challenges under pressure.
Ability to excel when tackling shifting daily priorities based on production model requirements.
Skill in debugging ambiguous, complex problems across multiple technology stack layers.
Clear communication and collaboration capabilities for working across time zones and during stressful incidents.
Dedication to refining engineering craft and caring about the responsible scaling and societal impacts of AI.
Minimum education requirement of a Bachelor’s degree or an equivalent combination of education, training, and experience in a relevant field.

Nice to Have

Previous background training large language models or working extensively with PyTorch, JAX, TPUs, or scale machine learning frameworks.
Contributions made to open-source large language model frameworks such as llm-foundry, open_lm, or mesh-transformer-jax.
Published research covering scaling laws, model training, or machine learning systems.
Familiarity with evaluation infrastructure, production machine learning systems, or observability tools.
Professional background as a quant, systems engineer, or other roles demanding operational excellence and technical depth.

Skills and Tools

PyTorch
LLMs
JAX
TPUs
Large-scale distributed systems
Production machine learning stacks

Salary and Benefits

Annual compensation ranging from £260,000 to £630,000 GBP.
Competitive compensation and benefits package.
Generous parental leave and vacation time.
Flexible working hours.
Optional equity donation matching.
Collaborative office environment in London.
Visa sponsorship provided where applicable.

About the Company

Anthropic functions as a public benefit corporation with the mission of building reliable, interpretable, and steerable artificial intelligence systems that benefit society. The organization operates as a single cohesive team tackling big science and empirical AI research, valuing long-term goals over smaller specific puzzles.

How to Apply

Candidates interested in this position should visit the official Anthropic careers page to submit an application.

Skills and tech stack

PyTorchLLM

Questions candidates ask

What is the location of the Research Engineer role?

The role requires working five days per week onsite in the London office.

What is the salary range for this position?

The annual salary range is £260,000 to £630,000 GBP.

Does Anthropic offer visa sponsorship for this job?

Yes, Anthropic sponsors visas for eligible candidates and partners with an immigration lawyer to assist with the process.

What level of experience is required?

The required years of experience correlate with internal job level requirements and minimum educational standards, requiring a Bachelor's degree or equivalent in a relevant field.

What does the work schedule look like during model launches?

During launches, the team works in tight lockstep, often putting in extended hours and responding to production issues during evenings and weekends.

Ready to apply?

The application is on the official Anthropic careers page.

Apply on company site