Math Device Extension Libraries Intern - 2027
Core
Design, develop, and optimize GPU-accelerated mathematical libraries (cuBLASDx, cuSolverDx, etc.) and Device eXtension (Dx) APIs for high-performance computing and AI applications.
Role type
Software engineering intern (high-performance computing)
Builds
GPU-accelerated math libraries and Dx APIs for NVIDIA's accelerated computing platform
Domain
High-performance computing, AI, GPU programming
Deliverable
production ML models
Required skills
C++ programming, CUDA/OpenMP/OpenCL/SYCL, linear algebra, FFT, data compression algorithms, performance analysis, test building
Preferred skills
Compiler implementation, GPU/network hardware architecture, Python, cuTile, CUDA Tile C++, LLMs, agentic workflows, RAGs, numerical methods, floating-point arithmetic
Technologies
CUDA, OpenMP, OpenCL, SYCL, Python, cuBLASDx, cuSolverDx, cuFFTDx, cuRANDDx, nvCOMPDx, NPPDx, libmathdx, MLIR
Responsibilities
Prototype and develop numerical algorithms for high-performance math device extension libraries; Analyze performance of GPU implementations and find opportunities for improvements; Prototype tooling to transform math libraries' APIs into LLM-friendly representations; Collaborate with team members to understand software use cases and requirements
Seniority
Intern
