Hybrid Linux HPC Architect for Scalable Infra

MBDA; Bolton; Hybrid; GBP 68,000 - 83,000; Full time

Benefits offered by this job

Bonus: up to 21% of base salary

Pension: up to 14% total contributions

Flexible working

Enhanced parental leave

Site facilities and parking

Healthcare Cash Plan

Job summary

MBDA in Bolton is seeking a forward-thinking HPC Architect to define the long-term roadmap for high-performance computing environments. You will shape storage, compute and interconnect strategies, aligning with MBDA's engineering objectives and international collaboration.

You will manage scalable HPC clusters, including CPU/GPU systems, and work with cross-functional teams to deliver high-throughput, reproducible workloads using SLURM, MPI and CUDA.

Qualifications

• Candidate must have expertise in HPC architecture and storage/compute design.
• Experience building scalable HPC environments for multi-national teams.
• Proficient in workload orchestration tools for HPC such as SLURM.
• Familiar with DV clearance processes and security standards.

Responsibilities

• Define long-term HPC architecture roadmap aligned with MBDA goals.
• Architect large-scale HPC clusters with CPU, GPU accelerators.
• Design high-performance networking and storage architectures.
• Provide technical leadership to HPC engineering and operations.
• Lead performance benchmarking and capacity planning.
• Guide software stack design including MPI, CUDA, libraries, containers.
• Ensure reproducibility and portability of workloads.

Skills

HPC architecture

Storage & interconnects

InfiniBand RDMA

High-throughput design

Team leadership

Tools

SLURM

Kubernetes for HPC

MPI

CUDA

Containers

Job description

MBDA in Bolton is seeking a forward-thinking HPC Architect to define the long-term roadmap for high-performance computing environments. You will shape storage, compute and interconnect strategies, aligning with MBDA's engineering objectives and international collaboration.

You will manage scalable HPC clusters, including CPU/GPU systems, and work with cross-functional teams to deliver high-throughput, reproducible workloads using SLURM, MPI and CUDA.