Senior Software Engineer, SRE and Production Engineering - DGX Cloud

at Nvidia

USD 184,000-287,500 per year

SENIOR

✅ On-site

Tech Stack
Tag name is followed by "@" symbol and proficiency level value. About proficiency levels:

1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;

3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;

7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;

10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.

AI Communication @ 6 Debugging @ 4 GPU @ 4 Go @ 7 IaaS InfiniBand @ 6 Kubernetes @ 4 Linux @ 4 NVLink @ 6 Networking @ 6 Observability @ 4 Python @ 7 SRE @ 4

Details

NVIDIA DGX Cloud builds and operates large-scale GPU infrastructure for AI workloads. The team is seeking software engineers with SRE or production engineering experience and hands-on experience with bare-metal NVIDIA systems. The team builds software and operational tooling that moves GPU capacity from installed hardware to production service in an IaaS production environment supporting BMaaS and VMaaS.

Responsibilities

Build automation for bare-metal provisioning, hardware validation, firmware and software upgrades, repair, and cluster lifecycle management.
Develop tools that interact with BMC and Redfish interfaces to monitor hardware health, manage server state, and support recovery workflows.
Handle and advance NVIDIA NVL72 systems and BlueField-3 or later DPUs across cloud partner and on-premises environments.
Diagnose failures across servers, DPUs, GPU systems, CPU systems, networking, Linux, and Kubernetes, and turn recurring issues into automated detection and repair.
Define validation and handoff criteria so new capacity enters production safely and consistently.
Participate in on-call duties, incident response, root-cause analysis, and follow-up work to implement permanent solutions.
Work with hardware, networking, platform, data center operations, and partner teams to resolve issues across ownership boundaries.

Requirements

8+ years of experience building software for or operating production infrastructure, including substantial hands-on bare-metal experience.
Strong Go or Python skills, with a record of delivering production automation and services.
Direct experience with BMC and Redfish for server provisioning, health inspection, power management, or fault diagnosis.
Practical experience working directly with NVIDIA GPU hardware, such as NVL72 systems, and BlueField-3 or newer DPUs.
Experience with Linux, firmware and driver management, network boot, and the server lifecycle from initial provisioning through repair.
Experience managing production reliability through on-call duties, incident response, observability, and durable solutions.
Ability to debug failures across hardware, host operating systems, networking, and distributed services.
Clear communication skills and demonstrated ownership of problems spanning multiple teams.
Bachelor's or master's degree in Computer Science, or equivalent practical experience.

Preferred Qualifications

Experience operating BlueField DPUs in DPU mode, including host-to-DPU connectivity and lifecycle debugging; equivalent DPU experience is also considered.
Background with NVLink, InfiniBand, Spectrum-X, or GPU cluster performance validation.
Experience building safe, repeatable workflows for rack-scale bring-up, firmware upgrades, hardware replacement, and customer handoff.
Experience with Kubernetes, GitOps, Argo CD, SLOs, and fleet-wide automation.

Compensation and Benefits

Base salary range: USD 184,000–287,500 per year.
Eligible for equity and benefits.

Applications will be accepted at least until October 16, 2026. NVIDIA is committed to fostering an inclusive work environment and is an equal opportunity employer.