AI Infrastructure Engineer: Scalable GPU Training on AWS

LinuxRecruit; Greater London; On-site; GBP 90,000 - 150,000; Full time

Job summary

LinuxRecruit in London is seeking a talented AI infrastructure engineer to own the systems powering large‑scale model training. You will architect and optimise distributed workloads across multi‑node GPU clusters in AWS, focusing on removing bottlenecks and maximising GPU utilisation.

You’ll collaborate with PyTorch‑based learning frameworks, Slurm, and Kubernetes, helping evolve the platform to support next‑generation GPU infrastructure and specialised compute providers, while shaping the

Qualifications

• Experience building and maintaining large-scale AI training platforms.
• Proficiency with GPU clusters, distributed training, and cost-aware scheduling.
• Strong knowledge of Linux, networking, and performance profiling.

Responsibilities

• Own the systems powering large-scale model training and optimise the training pipeline.
• Architect and manage distributed workloads across multi-node GPU clusters in AWS.
• Maximise GPU utilisation, remove bottlenecks and improve throughput.
• Manage cluster orchestration with Slurm and Kubernetes; evolve platform for future GPUs.
• Collaborate across teams to support PyTorch-based learning frameworks and storage/networking.

Skills

Distributed systems

GPU clusters

Slurm

Kubernetes

PyTorch

AWS

Performance optimisation

Tools

Slurm

Kubernetes

PyTorch

AWS

Job description

LinuxRecruit in London is seeking a talented AI infrastructure engineer to own the systems powering large‑scale model training. You will architect and optimise distributed workloads across multi‑node GPU clusters in AWS, focusing on removing bottlenecks and maximising GPU utilisation.

You’ll collaborate with PyTorch‑based learning frameworks, Slurm, and Kubernetes, helping evolve the platform to support next‑generation GPU infrastructure and specialised compute providers, while shaping the