Senior AI Systems Administrator: GPU HPC Infra Lead

CommonAI Holdings Ltd; Cambridge; Remote; GBP 65,000 - 90,000; Full time

Benefits offered by this job

Pension

Professional development

Networking opportunities

Office near Cambridge train station

Job summary

CommonAI CIC in Cambridge is seeking a Senior AI Systems Administrator to design, deploy and maintain multi-rack GPU server clusters for inference and training. You will work at the intersection of high-end hardware and software orchestration, ensuring scalable, reliable infrastructure for advanced AI research.

The role offers a hands-on environment, exposure to cutting-edge tooling, and opportunities to influence system performance, security and network design while collaborating with

Qualifications

• Significant experience as System Administrator or HPC/MLOps role.
• Experience with GPU server deployment and tuning.
• Advanced Linux and virtualization across distributions.
• Proficiency in Git and IaC tools (Terraform, OpenTofu).
• Strong network design, switch/router/firewall configuration.
• Ability to collaborate with users and stakeholders.

Responsibilities

• Provision and maintain multi-rack GPU clusters for inference/training.
• Tune and manage high-performance networking (Infiniband, RDMA, RoCE).
• Manage servers with hypervisors/VMs and multiple distros.
• Contribute to automation and infrastructure-as-code pipelines.

Skills

GPU server deployment

Linux administration

Networking technologies

Source code management

Infrastructure-as-code

HPC concepts

Collaboration with users

Tools

Git

Terraform

OpenTofu

Job description

CommonAI CIC in Cambridge is seeking a Senior AI Systems Administrator to design, deploy and maintain multi-rack GPU server clusters for inference and training. You will work at the intersection of high-end hardware and software orchestration, ensuring scalable, reliable infrastructure for advanced AI research.

The role offers a hands-on environment, exposure to cutting-edge tooling, and opportunities to influence system performance, security and network design while collaborating with