AI infrastructure engineer

LinuxRecruit; Greater London; On-site; GBP 90,000 - 120,000; Full time

Benefits offered by this job

Competitive salary

Equity in early-stage startup

London lab

Job summary

LinuxRecruit is seeking an experienced AI Infrastructure or MLOps Engineer to join our London lab. You will architect and optimize distributed training across multiple GPUs and machines in AWS, eliminate bottlenecks in the data path, and manage cluster orchestration with Slurm and Kubernetes.

The role requires deep PyTorch expertise, familiarity with transformer models, and experience deploying production AI systems.

Qualifications

• Proven track record building production AI systems.
• Strong PyTorch experience and understanding of transformers.
• Experience with distributed training and high-performance compute.

Responsibilities

• Architect and optimize distributed training across multiple GPUs and machines.
• Manage cluster orchestration using Slurm and Kubernetes; expand to specialised GPU providers.
• Collaborate across teams in a fast-moving startup to shape technology decisions.

Skills

Distributed training design

Performance optimization

Collaboration & ownership

Problem solving under pressure

Tools

PyTorch

Slurm

Kubernetes

AWS

Job description

Work on the cutting edge of AI.

This is an opportunity to join an AI startup working with the cutting edge in robotics at a stage where your impact will shape the future of the company itself. You will be joining a company that is still in stealth but has had a record breaking seed round of funding. And this isn't just another AI startup, this is a talent dense team of experts who are moving at an accelerated pace to solve the hardest problems in embodied AI.

You’ll be a deeply technical specialist, who will architect and optimize distributed training across multiple GPUs and machines in AWS. You will eliminate bottlenecks in the data path to ensure training is fast and as capital efficient as possible alongside managing cluster orchestration using slurm and Kubernetes while preparing to expand into specialised GPU providers. And finally you will master the stack from pytorch based learning libraries to complex data networking and GPU utilisation.

We're looking for an experienced AI Infrastructure or MLOps Engineer who has already built and operated production AI systems rather than someone at the beginning of their career. You'll have strong experience with the PyTorch ecosystem and a solid understanding of modern transformer architectures, including how to train and operate them at scale. Experience working with computer vision models and image-based datasets is particularly valuable, and we're looking for someone who thrives in an early-stage startup environment where ownership is high, pace is fast, and everyone contributes across the stack. This is a highly collaborative role, so you'll be excited to work alongside the team in the London- based lab every day.

In return, you'll join at a pivotal stage where your work will directly shape both the technology and the future of the company. You'll be building cutting-edge robotics AI, while benefiting from a competitive salary, meaningful founding equity, and the opportunity to work alongside a world-class team in a state-of-the-art London lab.