Senior HPC Cloud Engineer

Overview

• NVIDIA is looking for an exceptional engineer to grow and thrive alongside our HPC Infrastructure team that designs, evaluates, and optimizes the compute inrastructure powering NVIDIA's next-generation silicon design and AI workloads.
• You will be the primary architecture reviewer and performance champion for new data center clusters being built across multiple sites globally.
• What you'll be doing: In this position, you will own the build-out and day-to-day operation of NVIDIA's VLSI high-performance compute environments in the public cloud, and the repeatable patterns that let us stand up the next environment faster than the last.
• Bring up and operate large compute clusters on Google Cloud end to end - instance provisioning, machine images, OS pipelines, networking, identity and quota management.
• Build the automation that makes cluster bring-up reproducible: infrastructure-as-code, configuration management, image build pipelines, and host lifecycle automation including hostname and identity preservation across maintenance events.
• Extend the same patterns to new on-premises data centers so cloud and on-prem run on one toolchain rather than two, and work with networking and storage teams on cross-site connectivity, latency budgets, and what is cross-mounted versus kept site-local.
• This is a hands-on technical role.
• You will debug at depth when clusters misbehave - image and boot issues, mount failures, scheduler-host interactions, capacity and quota problems - and partner with the scheduling and operations engineers so what you build is operable by the whole team, not only by you.
• What we need to see: B.E./B.Tech/M.Tech in Computer Science, Electrical Engineering, or equivalent experience.
• 8+ years in infrastructure engineering, with substantial time running production compute at scale.
• Deep, hands-on expertise with a major public cloud, Google Cloud preferred - compute, networking, IAM, storage, quotas, and the operational realities of running thousands of instances rather than tens.
• Strong infrastructure-as-code and configuration management skills (Terraform, Ansible, Puppet, or equivalent), plus experience with machine image build and OS lifecycle pipelines for large fleets.
• Solid Linux systems administration and troubleshooting: filesystems, NFS mounts and latency, networking, boot and image pipelines.
• Working understanding of data center and network fundamentals for compute-intensive workloads - topology, bandwidth and latency characterisation, and the practical limits of cross-site operation - along with familiarity with batch schedulers (LSF, Slurm, or equivalent) and how compute nodes join, leave and are managed within them.
• Strong Python automation, experience bringing a new environment, cluster or data center online end to end, and the ability to work effectively across time zones.
• Semiconductor/EDA exposure is a plus, not a requirement.