Senior Math Libraries Engineer - Direct Sparse Solvers

Description

The team builds cuDSS, a CUDA library of direct solvers for sparse linear systems, used across scientific simulation, data analytics and AI on GPU-based data centers.

Its applications include computer aided engineering, electronic design automation, optimization, quantum chemistry, autonomous vehicles and large language models.

You will help write the algorithms and kernels behind these direct sparse solvers alongside other developers, and you should already have built and tuned accelerated computing kernels and want to push the field forward across several accelerated computing areas.

The company describes itself as an AI computing company, rooted in GPU hardware that dates back to 1999 and now powers deep learning, robotics and self-driving cars. It is looking for people who are creative, self-directed and enjoy hard problems.

Responsibilities

• Design, build and tune direct sparse solvers for current and upcoming GPU architectures.
• Work with library engineers, QA engineers and interns on every stage of library work, from design and implementation through testing, release and support.
• Partner with product management and internal and external stakeholders to define feature and performance needs and help shape the library technical roadmap.
• Identify chances to raise library quality, speed and maintainability in sparse linear algebra, including by re-architecting code and introducing new software practices.

Requirements

• A PhD or MSc in computer science, computational science and engineering, applied mathematics or a related science or engineering field, or equivalent experience.
• 5+ years building, debugging and optimizing high-performance numerical software in C++ with parallel programming. CUDA, MPI, OpenMP, OpenACC, pthreads or similar experience is preferred.
• Solid grounding in floating-point arithmetic, numerical analysis and sparse linear algebra primitives such as matrix-vector and matrix-matrix products and triangular solves.
• Experience developing, maintaining and testing scientific computing libraries.
• Strong collaboration, communication and documentation habits.

Conditions

Base salary depends on location, experience and other objective job-related factors. Expected base salary is 184,000 USD to 287,500 USD for Level 4 and 224,000 USD to 356,500 USD for Level 5.

This is an existing vacancy. Applications are accepted at least until November 3, 2026.

The company uses AI tools in its recruiting process.

The company is an equal opportunity employer and does not discriminate on protected characteristics, including in hiring and promotion.

Benefits

• Optional, preferred extras: knowledge of direct solver techniques such as reordering, multifrontal and supernodal factorizations, numerical pivoting and iterative refinement.
• Optional, preferred extras: solid knowledge of CPU or GPU hardware architecture and low-level GPU performance tuning.
• Optional, preferred extras: experience with modern engineering practices such as CI/CD systems, JIRA and AI agents.
• Optional, preferred extras: understanding of large-scale computing such as PDE solvers, eigenvalue solvers and time-domain simulation (for example CFD and FEA).
• Optional, preferred extras: experience working in a globally distributed, agile organization.