HPC Architect

In this role you shape the long-term HPC direction for a defence-focused environment, owning the architecture roadmap and ensuring the platform scales with engineering needs. You’ll lead the design of large, high-throughput clusters with CPU, GPU and accelerator-based compute, while shaping standards, performance, and interoperability across teams. Collaborate with engineers, vendors, and multinational stakeholders to deliver robust, future-ready HPC capability that supports mission-critical workloads. This is a senior, impact-driven role offering technical leadership and clear architectural ownership in a hybrid Bristol setting.

• base salary up to 75,000
• bonus of 20%+
• employer and employee pension contributions up to 14%
• enhanced parental leave up to 26 weeks

• Define and own the long-term HPC architecture roadmap.
• Architect scalable HPC environments covering compute, storage, interconnects, and supporting infra.
• Evaluate and select technologies across compute, storage, networking and cloud HPC.
• Design large-scale HPC clusters with CPU, GPU and accelerators for demanding workloads.
• Develop high-performance networking architectures (InfiniBand, RDMA, high-speed Ethernet).
• Define storage architectures across parallel file systems, object storage, and tiered solutions.
• Establish technical standards, reference architectures, and best practices for HPC.
• Lead benchmarking, capacity planning, workload modelling, and performance analysis.
• Identify bottlenecks and define improvements to optimise HPC performance.
• Provide technical leadership and architectural oversight to HPC teams.
• Collaborate with engineers, vendors and stakeholders to meet evolving requirements.
• Work with multinational teams to support capability development across the organisation.

• Strong Linux expertise.
• Proven experience designing or architecting High Performance Computing environments.
• Experience developing HPC architecture strategies, roadmaps, standards or reference architectures.
• Understanding of large-scale HPC clusters including CPU, GPU, accelerators.
• Knowledge of InfiniBand, RDMA and high-speed Ethernet.
• Experience designing storage using parallel file systems, object storage, and tiered storage.
• Experience performance engineering including benchmarking, capacity planning, workload modelling, bottleneck analysis.
• Knowledge of HPC workload scheduling/ orchestration (SLURM, Kubernetes).
• Familiarity with HPC software stacks (MPI, CUDA, compilers, libraries, containers).
• Understanding of workload reproducibility, portability, and performance in HPC environments.
• Ability to assess new compute/storage/network/cloud HPC technologies and translate options into long-term architecture decisions.
• Experience providing technical leadership or architectural guidance to teams.
• Strong stakeholder engagement and collaboration with distributed teams.

• Stakeholder engagement
• Collaborative mindset
• Leadership
• Linux mastery
• HPC architecture design
• InfiniBand, RDMA, high-speed Ethernet