CV/ML Platform Engineer
Allen Control Systems6 months ago
Austin, TX, USAMid Level
Responsibilities
- Deploy and operate bare-metal Kubernetes clusters hosting more than 130 NVIDIA GPUs, with hybrid burst capability to AWS.
- Manage NVIDIA GPU clusters for machine learning training.
- Own and maintain the ACS CV/ML CI/CD pipeline.
- Maintain model registration and versioning, experiment tracking, and model and data provenance systems.
- Improve ML model testing, performance analysis, and reporting tools.
- Automate model training and testing tasks to increase developer velocity.
- Coordinate CV/ML infrastructure with the broader Software Team platform infrastructure.
- Automate model optimization using TensorRT and quantization for deployment on NVIDIA Jetson and other edge hardware.
Requirements
- At least 2 years of experience in Platform Engineering or DevOps/MLOps.
- Strong programming skills for automating ML lifecycles and building custom CLI tools.
- Hands-on experience managing NVIDIA GPU infrastructure, CUDA libraries, development environments, GPU Operator, device plugins, and GPU scheduling.
- Experience implementing and maintaining MLOps platforms such as Kubeflow, MLflow, Weights & Biases, or DVC.
- Familiarity with high-performance storage solutions such as MinIO, WEKA, or Ceph and data orchestration for terabytes of video and image data.
- Experience building CI/CD pipelines with automated model validation, performance benchmarking, and artifact management for cloud and edge targets.
- Experience with TensorRT, ONNX, and quantization for cross-compilation to ARM targets such as NVIDIA Jetson.
- Proficiency with ELK, Prometheus, and Grafana observability stacks adapted for ML workloads.
- Strong Linux systems knowledge, including Debian or Ubuntu, high-throughput networking and storage, and security hardening for defense-grade production environments.
Benefits
- Health, Dental, and Vision Insurance
- Paid Time Off
Tech Stack
Categories
About Allen Control Systems
Allen Control Systems builds AI-enabled robotic weapon stations and small autonomous gun turrets that retrofit legacy and modern weapons for precise targeting, with emphasis on counter‑UAS missions. It sells hardware and software systems to U.S. defense and national security customers and integrators. The privately held company is headquartered in Austin, Texas, with an office in Alexandria, Virginia, and was founded by two former Navy electrical engineers.