General Atomics Aeronautical Systems, Inc. (GA-ASI) is looking to design, build, and maintain the compute, data, and CI/CD infrastructure that enables the development, test, and deployment of advanced perception algorithms for multi-sensor systems.
Requirements
- 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).
- Experience supporting AI/ML or perception pipelines for radar, EO/IR, or autonomy applications.
- Familiarity with C++, Python, and CUDA build environments.
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).
Other
- 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.
- Ability to obtain and maintain a DOD security clearance required.
- US Citizenship Required.
- Highly collaborative — able to work alongside algorithm developers, autonomy engineers, and IT security.
- Comfortable with rapid iteration, cross-functional coordination, and ownership of the end-to-end perception software lifecycle.
