About this role
Salary: £80,000 - 100,000 per year
Requirements: Significant experience as a Systems Administrator, HPC Engineer, MLOps Engineer, or in a similar roleHands-on experience deploying and tuning GPU serversStrong knowledge of InfiniBand, RDMA, and RoCE, with solid network design skillsAdvanced Linux, virtualisation, Git, and infrastructure-as-code experienceA collaborative approach and willingness to work closely with system usersBonus: Experience with parallel or high-availability storage such as Lustre, Ceph, Weka, or PEAK:AIO; Slurm or LSF; or Python scripting Responsibilities: Deploy, tune, and manage modern GPU servers across multiple racksBuild and operate high-performance networks using InfiniBand, RDMA, and RoCE, along with associated switches, routers, and firewallsRun Linux across multiple distributions, including Ubuntu and RHEL, on hosts running hypervisors and virtual machinesManage infrastructure as code using Git and Terraform or OpenTofuWork with researchers and engineers to continually improve system performance Technologies: AICephGitHardwareInfiniBandSupportLinuxMLOpsNetworkPythonRDMARHELTerraformUbuntuFabric More:
We are a mission-led, not-for-profit AI organisation in Cambridge. Startups, enterprises, the public sector, and academia share computing resources and engineering effort with us to build AI responsibly. This permanent or contract role supports multi-rack GPU clusters for training and inference, spanning hardware, networking, and orchestration. We offer access to the latest AI tools, a growing organisation where individual contributions make a visible difference, connections across industry, government, and academia, and a competitive package including pension, benefits, and professional development. Our city-centre office is a short walk from Cambridge station. Support is provided during UK business hours; occasional major maintenance is scheduled in advance and may take place on weekends. There is no on-call rota.
last updated 42 week of 2026