Hybrid HPC & AI Performance Engineer

Zerto; BLOOMINGTON (MN); Hybrid; USD 106,000 - 243,000; Full time

Job summary

Hewlett Packard Enterprise is seeking an HPC and AI Performance Engineer to benchmark, optimize, and analyze performance on HPC platforms. The role combines deep technical work with collaboration across Sales, Marketing, and Development to drive competitive benchmarking results.

The position requires strong C/C++/Fortran, MPI, OpenMP, CUDA/OpenACC experience, and Linux proficiency. A hybrid work model with some office days is expected to support collaboration.

Qualifications

• Experience benchmarking scientific applications and ISV modules.
• Ability to optimize benchmarks on Linux-based HPC systems.
• Strong written and verbal communication across teams.
• Willingness to work in a hybrid setting with occasional office presence.

Responsibilities

• Conduct scientific application performance benchmarking under tight deadlines.
• Optimize HPC benchmarks on HPE systems to support proposals.
• Develop understanding of HPC architectures and performance estimation methods.
• Analyze benchmark results to identify bottlenecks across compute, memory, interconnect, and I/O.
• Stay current on HPC products and optimization techniques.
• Collaborate with colleagues and external peers to achieve optimal benchmark performance.
• Provide benchmarking guidance to Sales, Marketing, and Development teams.

Skills

MPI parallelism

C/C++/Fortran

OpenMP

Python scripting

Linux environments

Profiling tools

CUDA/OpenACC

Education

Bachelor’s degree in technical field

Master’s or PhD preferred

Tools

MPI implementations

GPFS/Lustre file systems

CUDA GPUs

OpenACC

Job description

Hewlett Packard Enterprise is seeking an HPC and AI Performance Engineer to benchmark, optimize, and analyze performance on HPC platforms. The role combines deep technical work with collaboration across Sales, Marketing, and Development to drive competitive benchmarking results.

The position requires strong C/C++/Fortran, MPI, OpenMP, CUDA/OpenACC experience, and Linux proficiency. A hybrid work model with some office days is expected to support collaboration.