GPU Systems Engineer III (TS/SCI) – AI Clusters

Base-2 Solutions, LLC.; Bethesda, Northern (MD, KY); Hybrid; USD 140,000 - 200,000; Full time

Job summary

RPMGlobal in Bethesda, MD leads enterprise AI mission systems by designing, developing, and optimizing GPU clusters with emphasis on OS, hardware, GPU platforms, and high-speed networking in a secure government environment.

The role collaborates with AI/ML engineers to integrate GPUs with Linux-based systems, optimize drivers, analyze performance, and build tooling using Bash, Python, Ansible, Puppet, and Salt while maintaining comprehensive documentation and security compliance.

Qualifications

• Active TS/SCI with ability to obtain a CI Polygraph.
• Bachelor's degree with minimum six years of experience; three years may substitute for degree.
• Experience managing NVIDIA GPU data center platforms (DGX, HGX, H200, H100, L4s).
• Strong Linux expertise (RHEL, Ubuntu, Oracle, Rocky).
• DoD 8570.11 IAT Level II certification required; Level III acceptable.

Responsibilities

• Design, configure, and maintain GPU clusters.
• Collaborate with multidisciplinary teams to optimize architectures for performance and efficiency.
• Integrate GPUs with Linux-based systems for AI/ML workflows.
• Optimize GPU drivers for reliability and performance.
• Analyze GPU performance and bottlenecks; develop optimization strategies across hardware/software.
• Build and maintain debugging tools, profiling utilities, and performance analysis software for Linux.
• Leverage Bash, Python, Ansible, Puppet, and Salt for tooling and automation.
• Maintain technical documentation and Linux best practices.
• Support ATO activities and ensure compliance with federal security standards.

Skills

CI Polygraph

TS/SCI clearance

Linux expertise

Bash

Python

Ansible

Puppet

Salt

DoD 8570.11 IAT Level II

US citizenship

GPU cluster design

Education

Bachelor's degree

High School Diploma/GED

Associates Degree

Masters degree

PhD

Tools

NVIDIA DGX

NVIDIA HGX

NVIDIA H200

NVIDIA H100

NVIDIA L4S

Kubernetes

Argo

Airflow

Kubeflow

Prometheus

Grafana

Slurm

LSF

Job description

RPMGlobal in Bethesda, MD leads enterprise AI mission systems by designing, developing, and optimizing GPU clusters with emphasis on OS, hardware, GPU platforms, and high-speed networking in a secure government environment.

The role collaborates with AI/ML engineers to integrate GPUs with Linux-based systems, optimize drivers, analyze performance, and build tooling using Bash, Python, Ansible, Puppet, and Salt while maintaining comprehensive documentation and security compliance.