Research Engineer, Pretraining Scaling – London

Responsibilities

We are representing a confidential AI research organization seeking a Research Engineer, Pretraining Scaling to join their London team. This role sits at the intersection of research and engineering, where you'll own critical aspects of production model training at frontier scale.

• Own and optimize production pretraining pipelines, including model operations, performance tuning, observability systems, and reliability infrastructure
• Debug complex issues across the full technology stack—from hardware and networking layers through to training dynamics and evaluation frameworks
• Design and execute experiments to improve training efficiency, reduce step time, maximize uptime, and enhance model performance
• Respond to on-call incidents during model launches, diagnosing problems rapidly and coordinating cross-team solutions
• Build and maintain production logging, monitoring dashboards, and evaluation infrastructure
• Extend training codebases with new capabilities such as long context support or novel architectures
• Collaborate with distributed teams across time zones and coordinate with adjacent infrastructure and research teams
• Document systems, debugging approaches, and lessons learned to build institutional knowledge

What We're Looking For

Essential qualifications:

• Hands-on experience training large language models, or deep expertise with JAX, TPU, PyTorch, or large-scale distributed systems
• Genuine enthusiasm for both research and engineering work—candidates who thrive in a roughly 50/50 split between the two
• Excitement about on-call responsibilities, working extended hours during launches, and solving hard problems under pressure
• Ability to shift priorities daily based on production needs and what delivers the most impact
• Excellence at debugging complex, ambiguous problems across multiple technology layers
• Clear communication and effective collaboration skills, especially across time zones and during high-stress incidents
• Passion for refining your craft as a research engineer
• Interest in the societal impacts of AI and responsible development practices
• Bachelor's degree in a relevant field or equivalent combination of education and professional experience

Strong candidates may also have:

• Previous experience training LLMs or working extensively with JAX/TPU, PyTorch, or other ML frameworks at scale
• Contributions to open-source LLM frameworks
• Published research on model training, scaling laws, or ML systems
• Experience with production ML systems, observability tools, or evaluation infrastructure
• Background in systems engineering, quantitative roles, or positions requiring both technical depth and operational excellence

Role requirements:

• This position requires working in-office 5 days per week in London
• The role involves operational intensity including on-call duties and extended hours during model launches
• Visa sponsorship may be available; our client retains immigration counsel to support successful candidates

How We Work With You

Get A Job.ai represents this confidential client exclusively. Here's our process:

• Submit your application through the Get A Job.ai platform
• Our talent team will conduct an initial screening to understand your background and match with the role requirements
• Qualified candidates will be submitted to our client for consideration
• Please do not attempt to contact the client directly—all communication flows through Get A Job.ai to protect candidate and client interests

Pay

This role offers competitive compensation reflective of the high-impact nature of the work. Our client has indicated an annual salary range of £260,000–£630,000 GBP, along with optional equity participation and comprehensive benefits including generous vacation, parental leave, and flexible working arrangements.

Equal Opportunity: Get A Job.ai is committed to diverse representation in AI and technology. We encourage applications from candidates of all backgrounds and strongly urge you to apply even if you don't meet every listed qualification. Research shows that underrepresented groups are more prone to self-selection out of opportunities—we don't want to miss your application.