Senior Software Engineer, CUDA Python Core Libraries
Core
Design and implement Python APIs, bindings, and runtime infrastructure for NVIDIA's CUDA Core Libraries to enable GPU-accelerated software.
Role type
Senior Software Engineer (CUDA Python Core Libraries)
Builds
Pythonic APIs, language bindings, algorithms, and runtime infrastructure for CUDA
Domain
High-performance computing (HPC) and AI
Deliverable
production ML models | product features
Required skills
Python, C/C++, systems software concepts, API design, parallel programming, GPU programming, library development, testing, profiling, benchmarking
Preferred skills
CPU/GPU architecture, CUDA C++/Python, PyTorch, JAX, Numba, CuPy, Thrust, CUB, libcudacxx, LLVM, Clang, MLIR, Rust interoperability
Technologies
Python, C/C++, Rust, CUDA, Thrust, CUB, libcudacxx, LLVM, Clang, MLIR
Responsibilities
Design idiomatic Python APIs and bindings; Develop native C/C++ components; Define interoperability boundaries between Python, C/C++, Rust; Develop high-performance interfaces; Own features from design to maintenance; Improve developer experience via typing, packaging, and CI; Collaborate on shared architecture; Investigate performance and correctness issues
Seniority
Senior, hands-on IC