Senior Math Libraries Engineer - Sparse Linear Algebra

at Nvidia

USD 152,000-287,500 per year

SENIOR

✅ On-site

Tech Stack
Tag name is followed by "@" symbol and proficiency level value. About proficiency levels:

1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;

3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;

7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;

10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.

AI Agile @ 4 CI/CD @ 4 CUDA @ 4 Communication @ 7 Debugging @ 6 GPU @ 4 Jira @ 4 LLM MPI @ 4 Mathematics @ 4 Parallel Programming @ 6 Performance Optimization @ 4 Product Management @ 4 Project Management @ 4 Software Development @ 4

Details

We are looking for software engineers to join our development efforts in sparse linear algebra kernels for high-performance libraries such as cuSPARSE and cuDSS. These GPU-accelerated libraries and SDKs support AI, data analytics, scientific and engineering simulations, computer-aided engineering, electronic design automation, quantum chemistry, autonomous vehicles, large language models, and other applications.

Responsibilities

Design, implement, and optimize scalable, high-performance numerical sparse linear algebra software for existing and future GPU architectures.
Develop and optimize kernels for sparse BLAS operations, including matrix-vector and matrix-matrix products, direct sparse solvers, iterative sparse solvers, preconditioners, and algebraic multigrid.
Work with library engineers, QA engineers, and interns on topics ranging from sparse BLAS operations to advanced direct and iterative sparse solvers.
Collaborate with product management and internal and external partners to understand feature and performance requirements and contribute to technical roadmaps.
Improve library quality, performance, and maintainability through re-architecting and innovative software development practices.

Requirements

PhD or MSc degree, or equivalent experience, in Computational Science and Engineering, Computer Science, Applied Mathematics, or a related science or engineering field is preferred.
At least 5 years of experience developing, debugging, and optimizing high-performance sparse linear algebra software using C++ and parallel programming.
Experience with CUDA, MPI, OpenMP, OpenACC, pthreads, or equivalent technologies is preferred.
Strong fundamentals in floating-point arithmetic and implementation of sparse linear algebra primitives such as matrix-vector and matrix-matrix products.
Experience developing, maintaining, and testing sparse linear algebra libraries.
Strong collaboration, communication, and documentation skills.

Preferred Qualifications

Knowledge of CPU and/or GPU hardware architecture and low-level GPU performance optimization.
Familiarity with multifrontal factorizations, iterative solvers, preconditioners, and algebraic multigrid.
Experience adopting and advancing software development practices such as CI/CD, as well as using project management tools such as JIRA.
Understanding of large-scale computing technologies, including PDE solvers, eigenvalue solvers, and time-domain simulation methods such as CFD and FEA.
Experience working in a globally distributed and agile organization.

Benefits

Equity and benefits are provided.
NVIDIA is committed to fostering a diverse work environment and is an equal opportunity employer.

The base salary range is USD 152,000–218,500 for Level 3 and USD 184,000–287,500 for Level 4. The salary is determined based on location, experience, and the pay of employees in similar positions. Applications will be accepted at least until January 13, 2026. This posting is for an existing vacancy. NVIDIA uses AI tools in its recruiting processes.