Senior CUDA Math Libraries Engineer — Platform Bring-Up

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

Benefits offered by this job

Equity

Job summary

NVIDIA is seeking a Senior Software Engineer to advance CUDA Math Libraries on new and specialized platforms, overseeing integration, qualification, compliance, and performance readiness with the core library and devops teams.

You will work on high-performance numerical software in C++ and Python, lead platform bring-up across architectures, optimize HPC algorithms, and collaborate across CI/CD, testing, and documentation to ensure robust, scalable GPU software.

Qualifications

• PhD or MSc degree in Computational Science and Engineering, Computer Science, Applied Mathematics, or related science or engineering field (or equivalent experience).
• 5+ years of overall experience in developing high-performance numerical software.
• Experience with object-oriented programming, large system software architecture development, parallel computing, testing, maintenance, and performance optimization of HPC software using C++ and Python.
• Strong understanding of fundamental numerical methods and computations in science, engineering, and/or deep learning.
• Strong communication, collaboration, and documentation habits.
• Experience with software development practices, such as CI/CD systems, and project management tools such as JIRA.

Responsibilities

• Ensure platform readiness across Math Libraries (including cuBLAS, cuSPARSE, cuSOLVER, cuFFT, and others) for new and specialized platforms.
• Perform defect triage and isolation of platform-specific versus library-specific problems, fixing bugs to ensure functional correctness and referring to a library specialist as needed.
• Identify performance targets for new platforms, establish performance testing and fix performance regressions.
• Collaborate with CI/CD, build, and test infrastructure teams to establish platform-specific qualification and validation pipelines.

Skills

C++

Python

Parallel programming

HPC concepts

Software design

Education

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

Tools

JIRA

CI/CD pipelines

Job description

NVIDIA is seeking a Senior Software Engineer to advance CUDA Math Libraries on new and specialized platforms, overseeing integration, qualification, compliance, and performance readiness with the core library and devops teams.

You will work on high-performance numerical software in C++ and Python, lead platform bring-up across architectures, optimize HPC algorithms, and collaborate across CI/CD, testing, and documentation to ensure robust, scalable GPU software.