Remote GPU Math Libraries Intern - High-Performance APIs

NVIDIA Corporation; Polska; Hybrid; PLN 118,000 - 204,000; Full time

Job summary

NVIDIA Corporation is seeking a Math Device Extension Libraries Intern for 2027 to advance GPU-accelerated math libraries. You will work with senior engineers on cuBLASDx, cuSolverDx, cuFFTDx, cuRANDDx, nvCOMPDx, NPPDx and more, developing fast numerical algorithms on NVIDIA GPUs.

Based in Poland (Warsaw) with remote options, you will analyze performance, prototype tools, and collaborate across teams to transform APIs for AI and HPC workloads.

Qualifications

• Pursuing a MS or PhD in CS, AI, applied mathematics or related field.
• Strong C++ programming, debugging, performance analysis and test building skills.
• Experience with CUDA, OpenMP, OpenCL or SYCL.
• Collaborative, good communication and documentation habits.
• Familiarity with high-performance GPU computing and numerical methods.

Responsibilities

• Prototype and develop numerical algorithms for high-performance math libraries.
• Analyze GPU implementations and identify opportunities for optimization.
• Collaborate with teammates to understand use cases and requirements.
• Prototype tooling to translate APIs into LLМ-friendly representations across codegen tools.

Skills

C++ programming

Debugging

Performance analysis

Test building

Collaboration

Documentation

Education

MS or PhD in CS/AI/Applied Math

Tools

CUDA

OpenMP

OpenCL

SYCL

Job description

NVIDIA Corporation is seeking a Math Device Extension Libraries Intern for 2027 to advance GPU-accelerated math libraries. You will work with senior engineers on cuBLASDx, cuSolverDx, cuFFTDx, cuRANDDx, nvCOMPDx, NPPDx and more, developing fast numerical algorithms on NVIDIA GPUs.

Based in Poland (Warsaw) with remote options, you will analyze performance, prototype tools, and collaborate across teams to transform APIs for AI and HPC workloads.