Senior Software Engineer, SRE and Production Engineering - DGX Cloud
About this role
Accountabilities You will design, build, and maintain the software and operational tooling required to deploy, manage, and support large-scale GPU infrastructure. Your main responsibilities will include:
• Develop automation for bare-metal server provisioning, hardware validation, firmware and software upgrades, repairs, and cluster lifecycle management. • Build tools that interact with Baseboard Management Controller (BMC) and Redfish interfaces to monitor hardware health, manage server states, and support recovery workflows. • Operate and advance NVIDIA NVL72 systems and BlueField-3 or newer Data Processing Units (DPUs) across cloud partner and on-premises environments. • Diagnose and resolve complex failures involving servers, DPUs, GPU and CPU systems, networking, Linux, and Kubernetes. • Convert recurring infrastructure issues into automated detection, troubleshooting, and repair processes. • Define validation procedures and production handoff criteria to ensure new infrastructure capacity is introduced safely and consistently. • Participate in on-call rotations, incident response, root-cause investigations, and follow-up activities to implement permanent corrective solutions. • Collaborate with hardware, networking, platform engineering, data center operations, and partner teams to resolve issues across organizational boundaries. • Improve infrastructure reliability, operational consistency, and automation across the production fleet.
Requirements
The ideal candidate brings extensive software engineering and production infrastructure experience, with strong hands-on expertise in bare-metal systems and large-scale hardware operations. The following qualifications and capabilities are expected:
• At least 12 years of experience building software for or operating production infrastructure, including substantial hands-on bare-metal experience. • Strong programming skills in Go or Python, with a proven track record of delivering production automation and services. • Direct experience using BMC and Redfish interfaces for server provisioning, hardware health inspection, power management, or fault diagnosis. • Practical experience working directly with NVIDIA GPU hardware, particularly NVL72 systems, and BlueField-3 or newer DPUs. • Strong knowledge of Linux, firmware and driver management, network boot processes, and server lifecycle operations from initial provisioning through repair. • Experience maintaining production reliability through on-call responsibilities, incident response, observability, and implementation of durable solutions. • Ability to troubleshoot complex issues across hardware, host operating systems, networking, and distributed services. • Strong communication skills and demonstrated ownership of technical problems involving multiple teams. • Bachelor's or Master's degree in Computer Science, or equivalent practical experience.