Allen Control Systems
CV/ML Platform Engineer
About this role
Allen Control Systems, a defense startup developing autonomous targeting systems, seeks a CV/ML Platform Engineer to design and operate a 130+ GPU Kubernetes cluster infrastructure supporting computer vision and machine learning workloads. You'll own the ML platform stack including CI/CD pipelines, model management, and training infrastructure while ensuring seamless, high-volume model development.
What you'll do
- Deploy and operate bare-metal Kubernetes clusters with 130+ NVIDIA GPUs and AWS hybrid burst capability
- Manage CV/ML CI/CD pipelines and model deployment workflows
- Build and maintain ML infrastructure for model registration, versioning, experiment tracking, and data provenance
- Develop ML model testing, performance analysis, and reporting tools
- Optimize GPU cluster management for high-volume, low-friction model training
- Support CV/ML team with scalable compute and storage infrastructure
What they're looking for
- Kubernetes administration (bare-metal and cloud)
- NVIDIA GPU cluster management
- ML infrastructure and MLOps
- CI/CD pipeline development
- Python or systems programming
- AWS cloud services
- Data pipeline and model management tools
- Linux/distributed systems
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Allen Control Systems
Allen Control Systems builds autonomous defense systems and robotics platforms, developing real-time simulations, control software, and integrated hardware solutions. The company is hiring software engineers, systems integrators, field operations specialists, and quality engineers to support the development and deployment of next-generation autonomous defense technology.
View all jobs at Allen Control SystemsLikely interview questions
- Describe your experience managing large-scale GPU clusters. How have you optimized for high-volume model training with minimal friction?
- Walk us through a CV/ML CI/CD pipeline you've owned. What tools did you use and how did you handle model versioning and experiment tracking?