Principal System Software Engineer - CUDA Driver
Core
Architecting, implementing, and driving development of the CUDA driver, a core component for accelerating general-purpose computation on GPUs across diverse workloads including deep learning, scientific computation, and autonomous driving.
Role type
Principal System Software Engineer (CUDA Driver)
Builds
The CUDA platform and driver stack enabling GPU acceleration for AI, HPC, and visualization.
Domain
High-Performance Computing / GPU Systems Software
Deliverable
production ML models
Required skills
C programming, complex software stack design/debugging, OS interfaces (threads, process control, virtual memory), HW/SW co-design, performance modeling, multi-OS development, system-level architecture (interconnects, memory hierarchy), kernel mode development
Preferred skills
C++, system-level architecture (memory coherence, consistency models), driver programming for HW acceleration engines
Technologies
CUDA, C, C++, Linux/Windows (implied by multi-OS), GPU hardware
Responsibilities
Evangelize, architect, and implement new CUDA features; oversee and drive development efforts across multiple teams; collaborate with hardware architecture teams; define forward-looking improvements to CUDA APIs and programming model; design and maintain performance and precision modeling; write effective, maintainable, and well-tested code; develop code for multiple operating systems
Seniority
Principal, hands-on IC with strategic influence
