NVIDIA
Senior Software Engineer CUDA UMD - GPU Kernel Scheduling
$152K–$288K
US, CA, Santa Clara · Senior · Full-time · Posted Sep 1, 2026
- Experience
- Minimum 4 years of related development experience; multiple positions for varying experience levels are open.
- Education
- BS or MS in Computer Science, Electrical Engineering, or a related field (or equivalent experience).
About the job
Develop and improve NVIDIA’s CUDA Driver and programming models, including GPU scheduling features for AI/ML workloads.
What you'll do
- Architect and implement CUDA features, APIs, and programming-model improvements, including extending CUDA Graphs.
- Coordinate development across teams and develop code for multiple operating systems.
- Explore graph-based scheduling improvements for AI/ML workloads and write maintainable, well-tested code.
What they ask for
- Strong C and C++ programming skills.
- Experience driving projects across multiple teams and working with large codebases.
- Background with operating-system interfaces for threads, process control, and virtual memory; understanding of system architecture.
- Experience writing and debugging multithreaded programs; strong written communication and presentation skills.
Nice to have
- Prior parallel-computing experience, preferably writing CUDA programs or libraries.
- Knowledge of memory coherence and consistency models, kernel-mode development, Linux systems software, or programming-model support.
Pay
152,000 USD - 241,500 USD base salary (Level 3); 184,000 USD - 287,500 USD base salary (Level 4)
Tools
C, C++, CUDA, CUDA Graphs, Linux
AI focus
Explore using graphs to improve scheduling of AI/ML workloads on GPUs; the posting also states NVIDIA uses AI tools in recruiting.
Model-written summary · the original posting is authoritative