HPC Deployment Lead: Validation, AI & InfiniBand

NVIDIA; United States; On-site; USD 216,000 - 397,000; Full time

Benefits offered by this job

Equity

Job summary

NVIDIA is seeking an HPC Deployment Manager to supervise the deployment of InfiniBand and Ethernet technologies within the NVIDIA Infrastructure Specialists division. You will direct HPC service engineering, lead project planning and validation, and drive continuous improvement across AI HPC systems.

Ideal candidates have 10+ years in IT/HPC with 3+ years in management, deep knowledge of HPC interconnects, and strong scripting/Linux administration skills.

Qualifications

• 10+ years in IT/HPC or related field
• 3+ years in a management or leadership role
• Expertise in HPC systems design and planning
• Proficiency with low latency/high bandwidth interconnects (Infiniband, Ethernet)
• Experience with HPC system software cluster management tools and job schedulers (Slurm, salt, xCAT)
• Proficiency with shared memory parallelism (OpenMP, MPI, NCCL) and GPUs
• Strong scripting ability (Bash, Python, Perl) and programming fundamentals
• Administration of secure Linux/Unix OS (CentOS, Solaris)
• Ability to manage large, complex systems and troubleshoot network/infrastructure issues
• Bachelor's degree or equivalent experience in CS/IS or related field

Responsibilities

• Direct HPC service engineering functions for AI HPC systems
• Lead planning, implementation, and performance of HPC projects
• Develop and deploy procedures for system validation
• Drive tests and plans for HPC AI systems implementations
• Support strategy for service quality and continuous improvement
• Lead team development and career growth
• Build relationships with NVIDIA leaders, customers, partners
• Serve as domain authority with customers during planning to implementation

Skills

Leadership

HPC systems design

Scripting (Bash, Python, Perl)

Troubleshooting

Communication

Linux administration

Education

Bachelor's degree in computer science, information systems, or related field or equivalent experience

Tools

Slurm

xCAT

Salt

Ansible

Puppet

MPI

OpenMP

CUDA

Job description

NVIDIA is seeking an HPC Deployment Manager to supervise the deployment of InfiniBand and Ethernet technologies within the NVIDIA Infrastructure Specialists division. You will direct HPC service engineering, lead project planning and validation, and drive continuous improvement across AI HPC systems.

Ideal candidates have 10+ years in IT/HPC with 3+ years in management, deep knowledge of HPC interconnects, and strong scripting/Linux administration skills.