GM Performance Power Units - Concord, NCHPC Systems Engineer - OnsiteJob SummaryOwn day-to-day operation, maintenance, and optimization of production and development engineering HPC cluster(s) supporting various CAE, CFD, and FEA simulation suites for multiple cross-functional engineering groups.
Key Responsibilities- Operate and maintain RHEL-based NVIDIA Base Command Manager HPC environment. Responsible for patching, imaging, driver and firmware updates, and node health monitoring and remediation.
- Administer the Slurm scheduler: partitions, QoS, accounting, reservations, license tracking, prolog/epilog.
- Administer NFS storage volumes, quotas, and permissions.
- Administer license servers (FlexLM, RLM, DSLS) and license limit enforcements.
- Install and maintain commercial simulation software and associated environment modules (STAR-CCM+, ANSYS Fluent, Abaqus, Converge CFD, GT-Suite, Optistruct, JMAG Electromagnetic, etc).
- Provide direct user support: job script troubleshooting, failed and underperforming job diagnosis, user onboarding, documentation, and user training.
- Investigate performance and reliability problems to root cause using systematic diagnosis; benchmark and validate node and fabric performance.
- Write and maintain shell scripting for automation, health checks, reporting, and user-facing tooling.
- Own vendor trouble tickets (hardware, software, licensing) through to resolution.
- Consult and support build-out and commissioning of additional cluster capacity.
Required Qualifications- Linux system administration in a server or datacenter environment (RHEL family).
- Slurm or comparable HPC workload manager (PBS, LSF, Grid Engine).
- Cluster management tooling - Bright Cluster Manager / NVIDIA Base Command Manager.
- NVIDIA GPU and driver administration.
- Bash and python scripting.
- Strong networking fundamentals; RDMA/RoCE or InfiniBand exposure.
- MPI, NUMA topology, and CPU/memory performance tuning.
- NFS and shared filesystem administration.
- Familiarity with commercial CFD or FEA software.
- Methodical troubleshooter - forms a hypothesis, tests it, and works a problem to resolution.
- Comfortable with unfamiliar problems and vendor documentation.
- Clear communicator with engineering end users.
- Disciplined about change control on production systems.
It9s all part of our commitment to creating a workplace that9s as high-performing and forward-thinking as the people who make it succeed.
Join us and help power what9s next.You9ll play a pivotal role in ensuring the reliability and performance of a next-generation Formula 1 power unit. Our culture rewards precision, innovation, and the relentless pursuit of performance.
Only direct hires need apply to or inquire about job postings at GM Performance Power Units. We are not accepting calls, resumes or applications from recruiting firms at this time. 
