Senior Infrastructure Automation Engineer, Compute Platform

• BS or MS in Computer Science or equivalent experience.
• 6+ years in infrastructure engineering with strong config management depth - Ansible, Salt, Puppet, Chef, or comparable.
• Real experience with GitOps practice at scale: review workflow, environment promotion, drift detection, and safe rollback.
• Proficiency in Go, Python and shell, and comfort building the tooling rather than only using it.
• Experience automating stateful, long-lived infrastructure where you cannot simply destroy and recreate.

Ways to stand out from the crowd:

• You have brought a manually administered estate under configuration management while it stayed in production.
• Familiarity with LSF, Slurm, or another batch scheduler as a configuration target.
• CI/CD experience for infrastructure, including test environments that meaningfully resemble production.
• Observability instincts - you instrument what you automate.

Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 184,000 USD - 287,500 USD for Level 4, and 224,000 USD - 356,500 USD for Level 5.

Managing twenty-five scheduler cells by hand does not scale, and we are not going to try. NVIDIA is building a config-as-code foundation for its EDA compute farm, and we need an automation engineer to own it end to end. You are joining at the point where this is still partially manual and inconsistently applied across cells. That is the interesting version of the problem - you will be migrating off partial systems and reconciling drift, not starting from a blank repository.

What you’ll be doing:

• Designing and owning the configuration schema for LSF cell deployment, so that a policy change is written once, reviewed, tested, and applied identically everywhere it belongs.
• Building the deployment pipeline that takes scheduler configuration from merge to production across a federated estate, including staged rollout and rollback.
• Eliminating configuration drift across cells, and building the tooling that keeps it eliminated.
• Standing up the regression suite that lets us upgrade LSF on a schedule rather than on a dare.
• Working alongside our LSF internals engineer to encode hard-won scheduler knowledge into templates and policy, so that expertise lives in the repository instead of in one person’s head.