This is a fully remote position, open to applicants in California, +1 more state.

📋 Description

• Take full responsibility for the entire lifecycle of bare-metal GPU nodes, encompassing provisioning, delivery/onboarding, in-service operations, break-fix, and decommissioning across various regions.

• Develop and maintain automated, repeatable node-delivery pipelines to eliminate delivery backlogs and facilitate fleet expansion towards 10,000+ GPUs.

• Oversee DPU/SmartNIC and server firmware management, including BMC, BIOS, NIC, and GPU firmware version baselines, upgrades, and validations.

• Enhance fleet reliability by minimizing MTTR, leading incident response efforts, conducting root-cause analyses, and advancing hardware health monitoring.

• Engage in a sustainable 7×24 multi-region on-call rotation.

• Create runbooks and tools to minimize manual tasks.

• Collaborate with Storage/Image and Network teams to optimize the provisioning-to-handoff process.

• Establish standards for bring-up, rack, capacity, and acceptance for new GPU SKUs and data center regions.

⛳️ Requirements

• A minimum of 3 years of experience in large-scale bare-metal/server-fleet operations, HPC, or cloud infrastructure (6+ years for Senior-level candidates).

• Practical experience operating GPU servers at scale, such as NVIDIA HGX/DGX-class systems.

• Proficiency in driver/CUDA and firmware management.

• Strong skills in Linux systems.

• Experience with PXE/IPMI/Redfish, OS imaging, and automated provisioning.

• Familiarity with DPU/SmartNIC technologies, including NVIDIA BlueField, and bare-metal networking.

• Skills in infrastructure automation using Ansible, Terraform, Python, or Go.

• Comfortable taking ownership of on-call duties, incident management, and operational runbooks.

• Experience in multi-region/large-fleet operations is a significant advantage.