HPC / Linux Systems Administrator

VIRSIG, LLC•Stennis Space Center, MS

4d•$120,000 - $140,000•Onsite

About The Position

VIRSIG is looking for an experienced HPC/Linux Systems Administrator to handle on site system administration, operational support, and cybersecurity compliance for a Hewlett Packard Enterprise (HPE) High-Performance Computing cluster supporting Artificial Intelligence (AI) and Machine Learning (ML) workloads at NAVOCEANO, Stennis Space Center, MS. This system is a strategic Navy asset built on GPU accelerated compute nodes, an HPE Slingshot high speed interconnect, a Lustre parallel file system, and image based OS deployment infrastructure. If you enjoy solving complex technical challenges, working with cutting-edge HPC technologies, and supporting critical Department of Defense missions, we'd like to hear from you.

Requirements

U.S. citizen with an active DoD Top Secret clearance (Tier 5 investigation), adjudicated and in place before the contract start date
Meets the OPM qualification standard for IT Specialist (2210) or Computer Scientist (1550) at the GS-12 level or above, roughly 4 years of specialized IT/network administration experience
Bachelor's degree in Computer Science, or a bachelor's degree with 30 semester hours in math, statistics, and computer science (at least 15 of those hours in calculus or statistics), or equivalent experience
Hands on experience running Linux in an HPC environment
Experience with parallel file systems (Lustre), GPU accelerated compute nodes, and high speed interconnects (HPE Slingshot)
Experience with job schedulers (Slurm) and containers (Docker, Singularity/Apptainer)
Familiarity with DISA STIGs, vulnerability management, and DoD security frameworks (NIST 800-53, DoD Risk Management Framework)
Solid written and verbal communication skills, comfortable writing training materials and teaching government staff

Nice To Haves

Experience with HPE Cray System Management or HPE Performance Cluster Manager
Experience supporting AI/ML frameworks (TensorFlow, PyTorch) on HPC systems
Prior work on a DoD or Navy system administration contract, ideally in scientific or oceanographic computing
DoD 8570/8140 IA Technical Level I certification, or willingness to get it soon after hire

Responsibilities

Install, configure, and maintain HPC components, including CPU/GPU compute nodes, the Slingshot interconnect, and the Lustre parallel file system
Build and manage OS images for compute and service nodes so the software environment stays stable and consistent
Manage user accounts, access controls, and resource allocation, and run the Slurm job scheduler
Monitor system health and performance, handle routine maintenance and firmware updates, and coordinate hardware repairs with the vendor
Keep system documentation current, including configuration details, troubleshooting steps, and security protocols
Configure and manage the network, including switches, routers, and firewalls, and troubleshoot connectivity issues
Manage HPC storage (Lustre, HPE Parallel File System, GPFS), including backups and capacity monitoring
Install and maintain the HPC/AI-ML software stack (compilers, libraries, TensorFlow, PyTorch, Docker, Singularity/Apptainer) and manage licenses
Support users with performance tuning, GPU utilization, and getting the most out of the interconnect and file system
Harden systems to DISA STIG requirements and support quarterly STIG compliance checks and remediation
Run vulnerability scans and fix what they find, monitor security logs, respond to incidents, and help with A&A/ATO documentation
Train NAVOCEANO staff through on the job instruction and documentation, and carry out a knowledge transfer plan so government staff can run the system independently by the end of the contract
Help with monthly status reports, quarterly program reviews, and other NAVOCEANO briefings as needed