Senior HPC Dev Ops Engineer

Nvidia•Westford, MA

53d•$208,000 - $396,750

About The Position

As the HPC Operations Engineer for the new Accelerated Quantum Center located in Boston area, you'll be the linchpin in transforming next-generation HPC + Quantum infrastructure into robust, scalable architectures, balancing technical feasibility with customer needs. Engage with a team recognized for its industry leadership, advanced GPU supercomputers, and dedication to supporting our engineers, developers, and customers. You will help develop an innovative hybrid HPC + Quantum computing infrastructure featuring a substantial NVIDIA GB200 NVL72 GPU cluster (572 GPUs) that will be integrated with diverse quantum computing platforms. The lead HPC Engineer will offer technical mentorship, system administration, optimizing performance, and coordinating GPU compute and quantum system workloads.

Requirements

Proven experience (12+ years) in HPC systems engineering or administration within large-scale Linux-based GPU environments.
Extensive expertise in Slurm, Linux systems administration, and proficiency in configuration management tools like Ansible or Base Command (previously known as Bright Computing).
Practical familiarity with NVIDIA GPU technologies, InfiniBand, RDMA, and high-speed networking configurations.
Proficiency in containerization and orchestration tools like Singularity, Docker, or Kubernetes.
Knowledge of data center operations, including rack power, cooling methods (such as liquid-cooled systems), and network management.
Ability to automate, script, install/compile applications, and optimize performance.
Bachelor's degree in Computer Science, Electrical/Computer Engineering, Physics, or equivalent experience.
Outstanding problem-solving and diagnostic skills, and the ability to operate in a multidisciplinary, high-performance environment.

Nice To Haves

Exposure or familiarity of quantum computing systems (neutral-atom, trapped-ion, or superconducting).
Programming experience in Python, C/C++, or Shell Scripting for automation or performance tuning.
Knowledge of CUDA, MPI, and parallel programming paradigms is required.
Familiarity with NVIDIA HPC SDK, cuQuantum, or CUDA-Q toolchains.
Background supporting scientific or R&D computing environments.

Responsibilities

Build and operate a brand new hybrid compute environment spanning HPC and quantum systems.
Lead Linux provisioning, configuration management, and system tuning across hundreds of GPU nodes and supporting infrastructure.
Coordinate and optimize Slurm job scheduling - define policies, handle QoS, tune workloads, and help users translate research requirements into efficient batch workflows.
Coordinate data center tasks, partner with data center operations teams, connect with quantum lab.
Integrate and sustain container orchestration (e.g., Singularity, Docker, or Kubernetes for HPC) to back simulation workloads and quantum job processing.
Run storage environment consisting of Lustre, NFS, and Cloud storage.
Work closely with quantum engineering teams to merge quantum control nodes, orchestration gateways, and facilitate data exchange between HPC and quantum systems.
Addressing and improving performance for complex hybrid workloads, covering CUDA, MPI, and CUDA-Q applications.
Develop and automate operational workflows with Ansible, GitHub Actions, and CI/CD pipelines.
Support researchers and developers with environment setup, debugging, and performance profiling on NVIDIA hardware and quantum simulators as well as serve as the primary systems administrator and reliability owner for the GB200 GPU infrastructure.