Senior CUDA Math Libraries Engineer

NVIDIA; California; On-site; USD 152,000 - 288,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking a Senior Software Engineer to enable CUDA Math Libraries across new platforms from integration to performance readiness. You will work with core library and devops teams to ensure high-quality support across platforms, leveraging C++ and Python for HPC software development.

The role requires a PhD or MSc in a relevant field, 5+ years in high-performance numerical software, and strong experience with OO design, CUDA, and parallel programming.

Qualifications

• PhD or MSc in Computational Science, Computer Science, Applied Mathematics, or related field.
• 5+ years of experience in high-performance numerical software development.
• Strong OO programming and large-system software architecture experience with C++ and Python.
• Solid understanding of numerical methods and performance optimization for HPC.
• Excellent communication and documentation habits.
• Experience with CI/CD and Jira.

Responsibilities

• Enable CUDA Math Libraries (cuBLAS, cuSPARSE, cuSOLVER, cuFFT) on new platforms and ensure platform readiness.
• Triages defects, isolates library vs platform issues, and fixes bugs for correctness.
• Define performance targets and build validation pipelines with CI/CD and test infra.
• Collaborate with CI/CD, build and test teams to ensure qualification and platform bring-up.

Skills

C++

Python

HPC software

Parallel computing

OOP design

CI/CD

Education

PhD or MSc in Computational Science and Engineering, Computer Science, Applied Mathematics, or related field

Tools

JIRA

CI/CD pipelines

Job description

NVIDIA is seeking a Senior Software Engineer to enable CUDA Math Libraries across new platforms from integration to performance readiness. You will work with core library and devops teams to ensure high-quality support across platforms, leveraging C++ and Python for HPC software development.

The role requires a PhD or MSc in a relevant field, 5+ years in high-performance numerical software, and strong experience with OO design, CUDA, and parallel programming.