An SC-cleared HPC specialist is required to support a high-profile engineering programme within a globally recognised UK manufacturer operating across aerospace, defence, and mission-critical power systems.
The focus is modernising how HPC and HPC/ML workloads are built, packaged, and run using MPI-optimised container approaches - with pragmatic, implementable outcomes.
What you'll deliver:
* Rapid review of current HPC estate, workflows, and priority use cases (workshops + technical deep dives)
* Containerisation capability sessions focused on real HPC and HPC/ML workloads (not generic platform advice)
* Guidance on MPI‑ready container images, build tooling, and repeatable operational patterns
* Scheduler/runtime alignment and deployment approach (batch, interactive, data science workflows)
* Clear recommendations and next steps; produce a short written report where required
* Core technology focus
* MPI and performance‑aware containerisation (HPC‑first constraints)
* Apptainer / SIF images with OpenMPI (plus viable alternatives)
* Podman / Buildah tooling and secure build practices
Essential criteria (must have)
* Active SC Clearance
* Must meet programme nationality/eligibility requirements (sole nationality required; no dual nationality; no OCI)
* Hands‑on experience operating in Beowulf‑style cluster environments
* Excellent practical knowledge of Slurm and job execution patterns
* Strong understanding of best practices for MPI‑optimised container images (build, run, troubleshoot, performance)
* Ability to run structured interviews/workshops with customer engineering teams
* Awareness of SDLC considerations in research/HPC environments
* Familiarity with Jupyter (Hub/Lab/Notebooks) supporting data science workflows on shared compute
* Strong working knowledge of MPI patterns and real‑world operational constraints
* Hybrid Working
* Contract Initially 3‑Months with Lengthy Extensions
Nice to have:
* Singularity / SIF format and common HPC workflow patterns
* CFD tooling exposure (e.g., OpenFOAM)
* Docker and image lifecycle governance
* Kubernetes used as an HPC runtime environment (where appropriate)
* Run:AI and GPU slicing concepts
Location: Nottingham
#J-18808-Ljbffr