Join NVIDIA's High-Performance Computing (HPC) Infrastructure team as a Senior HPC Cloud Engineer. You will play a crucial role in designing, evaluating, and optimizing the compute infrastructure that powers NVIDIA's cutting-edge silicon design and AI workloads. This is an opportunity to be a primary architecture reviewer and performance champion for new global data center clusters.

This hands-on role involves building and managing NVIDIA's VLSI HPC environments within public cloud platforms, focusing on creating repeatable patterns for faster environment deployment. Responsibilities include end-to-end bring-up and operation of large compute clusters on Google Cloud, encompassing instance provisioning, machine images, OS pipelines, networking, identity, and quota management.

You will develop automation for reproducible cluster bring-up using infrastructure-as-code and configuration management. The role also extends to applying these patterns to new on-premises data centers, unifying cloud and on-prem toolchains. Collaboration with networking and storage teams for cross-site connectivity and latency optimization is key.

Troubleshooting complex issues like image/boot problems, mount failures, scheduler interactions, and capacity issues is part of the daily work. You will partner with scheduling and operations engineers to ensure operability across the team.

We are seeking an engineer with at least 8 years of experience in infrastructure engineering, specifically running production compute at scale. A deep, hands-on expertise with a major public cloud, preferably Google Cloud, is essential, covering compute, networking, IAM, storage, and quota management for thousands of instances.

Strong skills in infrastructure-as-code and configuration management (Terraform, Ansible, Puppet or similar) are required, along with experience in machine image build and OS lifecycle pipelines for large fleets. Solid Linux system administration and troubleshooting skills, including filesystems, NFS mounts, networking, and boot processes, are also necessary.

A working understanding of data center and network fundamentals for compute-intensive workloads is important, as is familiarity with batch schedulers (LSF, Slurm, or equivalent). Proficiency in Python automation and experience bringing new environments online end-to-end are crucial. While semiconductor/EDA exposure is a plus, it is not a mandatory requirement. A B.E./B.Tech/M.Tech in Computer Science, Electrical Engineering, or equivalent experience is expected.