Senior Staff Engineer, ML Ops (R4941)

Fdetimes; San Diego (CA); On-site; USD 200,000 - 290,000; Full time

Job summary

Shield AI is seeking a Senior Staff Engineer to help define and build the AI Factory Reference Architecture, a Kubernetes-native platform for AI development, distributed training, simulation, evaluation, and deployment. You will partner 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,

Qualifications

• Experience building Kubernetes-native AI or MLOps platforms.
• Deep understanding of modern AI training frameworks (PyTorch, Transformers).
• Experience operating GPU-accelerated infrastructure and distributed training systems.

Responsibilities

• Lead the design and implementation of an AI Factory Reference Architecture on Kubernetes.
• Partner with ML researchers to support training workflows and AI frameworks.
• Design self-service AI development workflows from local experiments to distributed execution.
• Build infrastructure for distributed training, simulation, inference, and RL workloads, with scalable orchestration and resource management.

Skills

Kubernetes-native AI

Distributed ML workloads

PyTorch

Transformers (Hugging Face)

Job description

Shield AI is a venture-backed defense-tech company with the mission of protecting service members and civilians with intelligent systems. Its products include Hivemind autonomy software, V-BAT and X-BAT aircraft, and Aechelon simulation and synthetic reality technologies. With offices and facilities across the U.S., Europe, the Middle East, and Asia-Pacific, Shield AI’s technology actively supports operations worldwide. For more information, visit www.shield.ai. Follow Shield AI on LinkedIn, X, Instagram, and YouTube.

Job Description

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.

What you'll do

• AI Platform Development: 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.
• AI Research Enablement: 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.
• Developer Experience: 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.
• Distributed AI Infrastructure: 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.
• Compute Platform: Design and optimize shared GPU infrastructure across cloud and on-premises environments. Improve resource utilization, scheduling efficiency, storage, networking, observability, and overall platform reliability.
• Data & Model Lifecycle: Build platform capabilities that enable dataset management, experiment tracking, artifact management, model versioning, evaluation, deployment, monitoring, and continuous model improvement.
• Platform Distribution: 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.
• Technology Leadership: Evaluate emerging AI infrastructure technologies and establish architectural patterns that balance scalability, performance, maintainability, and developer experience.
• Cross-Functional Collaboration: Work closely with AI researchers, autonomy teams, infrastructure engineers, and product teams to ensure the platform evolves alongside customer needs and advances in AI.

Key Outcomes

• Engineers move from idea to distributed training in hours instead of days.
• High GPU utilization through efficient scheduling with KAI on Kubernetes.
• Researchers use modern AI tooling without unnecessary platform friction.
• AI Workspaces become the standard development environment across autonomy programs.
• Training, simulation, evaluation, and deployment operate as a unified platform.
• The AI Factory Reference Architecture can be deployed consistently across cloud, on-premise, and air-gapped environments.
• Platform capabilities are reusable across multiple autonomy programs and customer deployments.

Required qualifications

• 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.