We are looking for a Senior to Staff Infrastructure Engineer to lead the design and evolution of large-scale, multi-GPU compute infrastructure used to train next-generation robotics and AI models. This role sits at the intersection of DevOps, MLOps, and distributed systems — owning architecture, reliability, and performance at scale in a fast-moving, cutting-edge environment.

Responsibilities

• Lead architecture and long-term technical direction of multi-GPU, cross-cloud training platforms
• Build and evolve infrastructure-as-code for provisioning, orchestration, and lifecycle management
• Architect and improve CI/CD systems for infrastructure and ML training workflows
• Optimize distributed training workloads — scheduling, resource utilization, observability
• Partner with ML engineers and researchers to enable efficient experimentation and productionization
• Mentor engineers and drive operational excellence across the org
• Document architecture, systems, and key technical decisions

Requirements

• Production-grade Kubernetes experience (CKA preferred)
• Hands-on Terraform (infrastructure-as-code)
• Kubernetes packaging and release management via Helm
• AWS cloud operations experience
• CI/CD pipeline experience including self-hosted runners (GitHub Actions)
• Prometheus/Grafana monitoring and alerting
• Linux administration, containerization, scripting (Python & Bash)
• Availability for on-call rotation

Nice to have

• GPU-accelerated Kubernetes clusters (NVIDIA)
• Cluster autoscaling (Karpenter)
• Workflow orchestration (Prefect)
• Gang scheduling, fair-share resource allocation