Role overview
Keep large-scale GPU and CPU infrastructure reliable and performant for production AI and high-performance computing workloads. This operations-focused SRE role spans bare-metal systems, Linux, networking, storage, and orchestration, with responsibility for validating new environments and improving how services are run. The position includes participation in a 24/7/365 on-call rotation and work with teams across infrastructure, platform engineering, tooling, and data centre operations.
Responsibilities
- Operate and troubleshoot production HPC infrastructure across compute, networking, storage, and orchestration layers.
- Test, benchmark, and assess new environments against reliability, scalability, and performance expectations before production use.
- Tune Linux systems and hardware, including kernels, firmware, and storage subsystems, for demanding workloads.
- Maintain bare-metal systems using out-of-band management tools and improve automation and operational tooling.
- Lead incident response and root-cause analysis, then implement corrective actions and reduce repetitive work.
- Improve monitoring, telemetry, alerting, and operational processes in partnership with engineering teams.
Requirements
- Extensive experience operating large-scale distributed infrastructure, including recent hands-on work with HPC or AI compute environments.
- Strong Linux administration and production troubleshooting skills.
- Experience with bare-metal systems, performance tuning, and hardware management tools such as IPMI, iLO, iDRAC, or Redfish.
- Solid networking fundamentals; familiarity with GPU computing and high-performance fabrics such as InfiniBand or RoCE.
- Experience with infrastructure automation or scripting, observability systems, workload schedulers, and parallel storage.
- Strong incident ownership, communication, and collaboration skills; experience with incident, problem, and change management.
Nice to have
- Experience with GPU-accelerated HPC at scale, HPC benchmarking, or multi-site infrastructure.
- Familiarity with Kubernetes-based environments, LPIC certification, or ITIL Foundation (or equivalent experience).
- A degree in computer science, engineering, or a related field, or equivalent practical experience.
Benefits and work setup
- Work with advanced GPU and AI infrastructure in a globally distributed engineering environment.
- The source highlights opportunities for technical growth, learning, collaboration, and operational ownership.