Senior HPC Applications Engineer: Performance Focus

Next Silicon Ltd; Minneapolis (MN); Hybrid; USD 120,000 - 180,000; Full time

Job summary

NextSilicon is seeking an experienced Staff Applications Engineer to advance HPC applications onto new platforms and measure performance at scale. You will analyze customer-defined workloads, identify bottlenecks, and contribute to software features for a novel dataflow architecture.

You will work on a multidisciplinary team across graph algorithms, weather prediction, seismic imaging, genomics, and molecular dynamics, with a strong emphasis on MPI/OpenMP parallelism, C/C++, and PyTorch.

Qualifications

• Bachelor of Science in an engineering discipline with a significant focus on scientific computing
• Masters or Ph.D. with relevant specialization is a plus
• A strong parallel programming experience using both MPI and OpenMP is required
• Hands-on experience with development in one or more HPC application domains
• Hands-on experience with one or more AI/ML frameworks, such as PyTorch
• Proficiency in C/C++ and/or Fortran
• Ability to measure application-level performance and profile HPC applications at the node level and at scale
• Expertise in competitive performance analysis is strongly preferred
• Willingness to travel as necessary
• US citizenship with eligibility to visit US government research facilities

Responsibilities

• Profile and identify bottlenecks of a wide range of HPC applications on different architectures
• Develop creative algorithmic and software solutions to solve bottlenecks and accelerate applications on a novel dataflow architecture
• Translate application requirements into software features and hardware requirements
• Develop performance models for future hardware architectures

Skills

MPI

OpenMP

C/C++

Fortran

Education

Bachelor of Science in engineering (scientific computing focus)

Master or PhD with relevant specialization (plus)

Tools

PyTorch

Job description

NextSilicon is seeking an experienced Staff Applications Engineer to advance HPC applications onto new platforms and measure performance at scale. You will analyze customer-defined workloads, identify bottlenecks, and contribute to software features for a novel dataflow architecture.

You will work on a multidisciplinary team across graph algorithms, weather prediction, seismic imaging, genomics, and molecular dynamics, with a strong emphasis on MPI/OpenMP parallelism, C/C++, and PyTorch.