Perception MLOps Infrastructure Engineer jobs in United States
cer-icon
Apply on Employer Site
company-logo

General Atomics · 2 months ago

Perception MLOps Infrastructure Engineer

General Atomics Aeronautical Systems, Inc. (GA-ASI) is a world leader in remotely piloted aircraft and tactical reconnaissance radars. The Perception MLOps Infrastructure Engineer will design, build, and maintain the infrastructure that supports advanced perception algorithms for multi-sensor systems, ensuring scalable and reproducible environments for various workloads.

EnergyEnergy EfficiencyManufacturingWireless
badNo H1BnoteSecurity Clearance RequirednoteU.S. Citizen Onlynote

Responsibilities

Architect, deploy, and maintain on-prem and isolated network compute infrastructure supporting perception algorithm development and test (GPU servers, storage arrays, and networked development hosts)
Design and manage GitLab CI/CD pipelines for build, test, and container deployment of perception software baselines (C++, CUDA, Python, ROS2)
Support algorithm developers with containerized and reproducible environments for ML training, sensor simulation, and embedded inference (Docker, Podman, Singularity)
Implement and maintain infrastructure-as-code for provisioning and configuration management (Ansible, Terraform, or equivalent)
Manage integration of data management tools (DVC, MLflow, Git LFS) for large datasets, model artifacts, and version tracking
Ensure secure network configuration and compliance with NIST SP 800-171 and corporate cybersecurity controls
Optimize GPU cluster scheduling and resource utilization (e.g., Slurm, Kubernetes, or GitLab runners for H100-class nodes)
Collaborate closely with perception algorithm engineers, autonomy software leads, and IT security to deliver reliable, high-throughput development pipelines
Support integration and test of perception software in hardware-in-the-loop and flight test environments

Qualification

Linux system administrationGitLab CI/CDContainerization (Docker/Podman)GPU compute environmentsSource control managementKubernetesNIST cybersecurity frameworksC++PythonSystem thinkerDocumentation skillsHighly collaborative

Required

Typically requires a bachelors, masters degree or PhD in computer science, engineering, mathematics, or a related technical discipline from an accredited institution and progressive machine learning engineering experience as follows; five or more years of experience with a bachelors degree or three or more years of experience with a masters degree. May substitute equivalent machine learning engineer experience in lieu of education
Proficiency with Linux system administration, networking, and shell scripting
Experience with GitLab CI/CD or comparable build automation systems
Strong working knowledge of containerization (Docker/Podman) and environment reproducibility for development and deployment
Familiarity with GPU compute environments (CUDA drivers, Slurm scheduling, NVIDIA management tools)
Demonstrated experience maintaining source control and artifact management systems (Git, DVC, Artifactory)
Excellent documentation and troubleshooting skills across heterogeneous systems
Ability to obtain and maintain a DOD security clearance required

Preferred

Experience supporting AI/ML or perception pipelines for radar, EO/IR, or autonomy applications
Familiarity with C++, Python, and CUDA build environments
Experience in air-gapped or classified network environments
Knowledge of Kubernetes, MLflow, or Prometheus/Grafana monitoring
Understanding of DoD cybersecurity frameworks (RMF, NIST 800-171, STIG compliance)
Prior experience in aerospace, defense, or autonomy systems integration

Company

General Atomics

company-logo
General Atomics is a defense and technology company specializing in research and technology development.

Funding

Current Stage
Late Stage
Total Funding
$6.76M
Key Investors
US Department of Energy
2023-08-14Grant
2023-05-09Grant
2022-09-19Grant

Leadership Team

leader-logo
Wayne Solomon
Vice President, Magnetic Fusion Energy
linkedin
Company data provided by crunchbase