Senior Infrastructure Engineer

Description

Quantum Sky is seeking a Senior Infrastructure Engineer to design, build, and operate the cloud and AI infrastructure behind Project 137, the company's applied-research and IP engine. Reporting to the Director of Infrastructure Engineering, this engineer will own Google Cloud Platform (GCP) and Kubernetes environments that run AI/ML training, inference, and data workloads.

The role is approximately 85% hands-on engineering and 15% project coordination. The ideal candidate is a senior practitioner with deep GCP expertise, production experience supporting AI workloads, and the organizational discipline to plan and schedule infrastructure work across teams.

Responsibilities:

• Design, build, and operate production infrastructure on Google Cloud Platform, including landing zones, VPC networking, Cloud IAM, storage, and organization policy.
• Engineer and scale infrastructure for AI/ML workloads: GPU and TPU capacity, distributed training, model serving and inference, and high-throughput data pipelines.
• Support data-science and research teams on Vertex AI, Kubeflow, and related MLOps tooling, including environment provisioning, model registries, and experiment infrastructure.
• Build and maintain hybrid connectivity between on-premises data-center and cloud environments, including Cloud Interconnect, VPN, DNS, and load balancing.

• Architect and operate GKE clusters, including GPU node pools, autoscaling, workload identity, networking and ingress, and version upgrades.
• Build reusable platform components and Helm charts that give engineering and research teams self-service access to compute and AI tooling.
• Define and automate infrastructure with Terraform and Git Ops practices, using Git as the source of truth for configuration, manifests, and CI/CD pipelines.

• Implement observability, alerting, and SLOs using Cloud Operations, Prometheus, and Grafana; participate in on-call and lead root-cause analysis on incidents.
• Apply cloud security controls: least-privilege IAM, secrets management, container image scanning, policy-as-code, and software supply-chain protections.
• Plan, build, and test backup and disaster-recovery capabilities for platform and AI workloads.
• Mentor engineers, lead design reviews, and document architecture standards and runbooks.

• Plan and schedule infrastructure work streams, maintaining timelines, milestones, and dependency tracking across engineering, research, and operations teams.
• Organize technical work into clear tasks and priorities, and report progress and blockers to the Director of Infrastructure Engineering.

Required:

• Education: Bachelor's degree in computer science, engineering, or a related technical field. Equivalent experience considered.
• Experience: 5+ years in infrastructure, platform, cloud, or site reliability engineering, including 2+ years of hands-on production work on Google Cloud Platform.
• Google Cloud: Deep expertise across Compute Engine, GKE, VPC, Cloud IAM, Cloud Storage, Cloud Operations.
• AI infrastructure: Experience supporting AI/ML workloads in production, including GPU or TPU provisioning, training and inference platforms, and MLOps tooling such as Vertex AI or Kubeflow.
• Kubernetes: Production experience operating Kubernetes, including cluster operations, Helm, networking and ingress, RBAC, autoscaling, and upgrades.
• Automation: Strong Terraform and Git-based workflow skills; experience with CI/CD tooling such as Cloud Build, Git Hub Actions, Git Lab CI, or Argo CD.
• Scripting: Proficiency in Python, Go, or Bash for automation and tooling.
• Networking and systems: Solid understanding of Linux, TCP/IP, DNS, load balancing, and hybrid cloud connectivity.
• Organization: Proven ability to plan, schedule, and track technical work across multiple concurrent work streams using tools such as Jira or Confluence.
• Communication: Clear written and verbal skills; able to explain technical trade-offs to engineers, researchers, and leadership.

Desired:

• Google Cloud Professional Cloud Architect, Professional Cloud Dev Ops Engineer, or Professional Machine Learning Engineer certification.
• Certified Kubernetes Administrator (CKA), Application Developer (CKAD), or Security Specialist (CKS).
• Experience with GPU scheduling and sharing on Kubernetes (NVIDIA GPU Operator, Kueue, or Ray) and large-scale distributed training.
• Familiarity with data-center fundamentals for high-density compute,…