Senior SRE for GPU Cloud: Kubernetes, Observability

NVIDIA; California (MO); On-site; USD 168,000 - 334,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA DGX Cloud is building and operating large-scale GPU infrastructure for AI workloads. We seek a Senior Reliability Engineer to drive automation, tooling, and production systems across Kubernetes-based clusters and DGX environments.

You will develop observability stacks, define SLOs/SLIs, and partner with cross-functional teams to ensure production readiness. A strong background in SRE, distributed systems, and cloud-native ops is required.

Qualifications

• 8+ years building or operating production infrastructure.
• Strong programming skills in Python, Go, or similar.
• Expert-level knowledge with Linux, Kubernetes, containers, cloud infrastructure, or infrastructure automation.
• Solid grasp of SRE principles, such as SLOs, SLIs, error budgets, and incident management.
• Ability to troubleshoot distributed systems in production.
• Experience building and operating observability stacks (monitoring, logging, tracing).
• Clear communication and ability to work across teams.
• BS/MS in Computer Science or equivalent experience.

Responsibilities

• Build and operate automation for large-scale Kubernetes clusters across NVIDIA Cloud Partners (NCP) and on-prem environments.
• Develop tools and services for provisioning, validation, upgrades, monitoring, repair, and cluster lifecycle operations.
• Improve Day 0 / Day 1 / Day 2 workflows for cluster bringup, handoff, and production operations.
• Define SLOs/SLIs, monitor error allowances, and streamline reporting
• Reduce manual production touches through APIs, GitOps, automation, and agent-assisted workflows.
• Participate in on-call, incident response, debugging, and durable follow-up work.
• Partner with platform, storage, networking, security, and workload teams to make infrastructure production-ready.

Skills

Python

Go

Linux

Kubernetes

CloudInfrastructure

SREPrinciples

DistributedSystems

Observability

Education

BS/MS in CS or equivalent

Tools

OpenTelemetry

Prometheus

Grafana

ELK Stack

Lightstep

Splunk

Terraform

ArgoCD

Job description

NVIDIA DGX Cloud is building and operating large-scale GPU infrastructure for AI workloads. We seek a Senior Reliability Engineer to drive automation, tooling, and production systems across Kubernetes-based clusters and DGX environments.

You will develop observability stacks, define SLOs/SLIs, and partner with cross-functional teams to ensure production readiness. A strong background in SRE, distributed systems, and cloud-native ops is required.