Role Overview
NVIDIA's accelerated computing platform has revolutionized HPC and AI. Our team develops GPU-accelerated mathematical libraries. The Dx APIs allow developers to embed highly-optimized mathematical operations. You will design, develop, and optimize math libraries for the future.
Responsibilities
- Prototype and develop numerical algorithms for high-performance math device extension libraries: cuBLASDx, cuSolverDx, cuFFTDx, cuRANDDx, nvCOMPDx, NPPDx or libmathdx on ground breaking NVIDIA graphics processors.
- Analyze the performance of GPU implementations and find opportunities for improvements
- Collaborate with team members to understand software use cases and requirements
- Prototype tooling and solutions to help transform math libraries’ APIs into LLM-friendly representations across popular codegen tools.
Requirements
- Pursuing a MS or PhD degree in Computer Science, Artificial Intelligence, Applied Mathematics, or related field
- Strong C++ programming, debugging, performance analysis, and test building skills
- Parallel or GPU programming experience using CUDA, OpenMP, OpenCL, SYCL or similar
- Experience with linear algebra or FFT or data compression algorithms
- Collaboration, communication, and documentation habits.
Nice to Have
- Understanding of composability and fusions, compilers, implementation of programming languages
- Knowledge of GPU and network hardware architecture
- Familiarity with Python and recent CUDA developments like cuTile and CUDA Tile C++
- Prior work in AI-assisted software development, knowledge of LLMs, MCP, and building RAGs or agenting tooling
- Exposure to numerical methods, floating-point arithmetic and numerical error analysis