On-Site HPC & GPU Infrastructure Engineer

Base-2 Solutions; Bethesda (MD); On-site; USD 120,000 - 150,000; Full time

Job summary

Base-2 Solutions seeks a mid-career Systems Engineer specializing in HPC and GPU infrastructure to design, develop, and optimize GPU clusters for Intelligence Community customers. This is a 100% on-site position at the customer site in Bethesda, Maryland.

You will install and maintain on-premises and cloud GPU/HPC hardware, analyze performance, configure Slurm, PBS, Airflow, and Kubernetes, and contribute to testing, documentation, and Linux-based GPU optimization.

Qualifications

• Bachelor's degree in CS/EE with 4+ years of relevant systems engineering experience.
• Proficiency in Linux OS integration and hardware architecture.
• Experience with scripting (Python, Bash) and automation tools.
• U.S. Citizenship and DoD 8570 IAT Level II/III certs required.

Responsibilities

• Install and maintain on-premises and cloud GPU/HPC hardware.
• Analyze cluster performance and optimize Linux applications.
• Configure Slurm, PBS, Apache Airflow, and Kubernetes workload managers.
• Develop power-management techniques to optimize GPU consumption.
• Document architectures and Linux best practices for GPU development.
• Monitor industry trends and propose Linux-based GPU optimizations.
• Share regular technical updates with the team.

Skills

Problem solving

Linux proficiency

Scripting (Python)

Communication

Teamwork

Education

Bachelor's degree in CS/EE

Tools

Ansible

Puppet

Salt

Terraform

Docker

Kubernetes

Prometheus

Grafana

Job description

Base-2 Solutions seeks a mid-career Systems Engineer specializing in HPC and GPU infrastructure to design, develop, and optimize GPU clusters for Intelligence Community customers. This is a 100% on-site position at the customer site in Bethesda, Maryland.

You will install and maintain on-premises and cloud GPU/HPC hardware, analyze performance, configure Slurm, PBS, Airflow, and Kubernetes, and contribute to testing, documentation, and Linux-based GPU optimization.