Senior Staff Engineer, ML Ops (R4941)

Shield AI•San Mateo, CA

2d•$280,000 - $420,000

About The Position

Shield AI builds autonomy systems for defense applications, including air, maritime, and space platforms operating in complex and contested environments. We are building the AI Factory Reference Architecture, a Kubernetes-native platform for developing, training, evaluating, and deploying next-generation AI systems. The AI Factory serves two purposes. Internally, it powers autonomy development across Hivemind and other AI programs. Externally, it becomes the reference architecture deployed into customer environments, spanning commercial cloud, on-premise infrastructure, sovereign deployments, and fully air-gapped systems. We are looking for a Senior Staff Engineer to help define and build this platform. You will partner closely with ML researchers, platform engineers, and autonomy teams to deliver an exceptional developer experience for training and deploying modern AI models. Success in this role requires balancing researcher productivity, platform simplicity, operational excellence, and long-term maintainability. You will work hands-on across the stack, helping shape both the platform architecture and its implementation while staying closely aligned with the rapidly evolving AI ecosystem.

Requirements

Experience building Kubernetes-native AI or MLOps platforms supporting distributed machine learning workloads.
Deep understanding of modern AI training frameworks, including PyTorch, Hugging Face Transformers and distributed training techniques.
Experience operating GPU-accelerated infrastructure and distributed training systems.
Strong understanding of Kubernetes, Linux, networking, security, storage, and distributed systems.
Experience with GPU scheduling concepts and large-scale AI workloads.
Experience packaging and deploying cloud-native infrastructure using Terraform and Helm.
Strong software engineering skills in Python and Golang and modern cloud-native technologies.
Experience collaborating closely with ML researchers to translate research workflows into scalable platform capabilities.

Nice To Haves

Experience with Ray or other distributed AI orchestration frameworks.
Experience with KAI, Slurm or other GPU scheduling technologies.
Experience supporting reinforcement learning, simulation-driven training, robotics, or autonomy workloads.
Experience deploying and optimizing AI models for edge hardware.
Experience designing infrastructure for classified, sovereign, or air-gapped environments.
Experience with observability technologies such as OpenTelemetry, Prometheus, and Grafana.
Experience contributing to or maintaining open-source infrastructure projects.

Responsibilities

Lead the design and implementation of the AI Factory Reference Architecture, delivering a Kubernetes-native platform for AI development, distributed training, simulation, evaluation, and deployment.
Partner directly with ML researchers to understand evolving training workflows and ensure the platform supports state-of-the-art AI frameworks, foundation model development, reinforcement learning, distributed training, and emerging research workflows.
Design self-service AI development workflows that enable engineers to move seamlessly from local experimentation to large-scale distributed execution using familiar open source tools and frameworks.
Build the infrastructure required to support distributed training, simulation, inference, and reinforcement learning workloads. Evaluate and integrate orchestration, scheduling, and resource management technologies to maximize scalability and developer productivity.
Design and optimize shared GPU infrastructure across cloud and on-premises environments. Improve resource utilization, scheduling efficiency, storage, networking, observability, and overall platform reliability.
Build platform capabilities that enable dataset management, experiment tracking, artifact management, model versioning, evaluation, deployment, monitoring, and continuous model improvement.
Develop repeatable deployment and lifecycle management solutions using Infrastructure as Code and modern platform engineering practices. Support commercial cloud, customer-managed infrastructure, sovereign environments, and fully air-gapped deployments.
Evaluate emerging AI infrastructure technologies and establish architectural patterns that balance scalability, performance, maintainability, and developer experience.
Work closely with AI researchers, autonomy teams, infrastructure engineers, and product teams to ensure the platform evolves alongside customer needs and advances in AI.