Join NVIDIA, the pioneers of the GPU revolution, and contribute to the future of AI and parallel computing. We are seeking experienced software engineers to develop the CUDA driver for Tegra, the platform that powers advanced computation across diverse workloads including deep learning, scientific research, autonomous vehicles, and gaming. This role involves shaping the unified programming model that connects NVIDIA hardware to a wide array of computational tasks, requiring deep interaction with GPU hardware, kernel and user mode drivers, and the operating system.

Architect and implement innovative features for the CUDA driver, driving the evolution of NVIDIA's compute platform. Coordinate development efforts across multiple teams to ensure seamless integration and delivery. Contribute to defining future improvements for CUDA APIs and the programming model. Produce high-quality, maintainable, and thoroughly tested code, with a specific focus on Tegra devices. Integrate AI best practices into development workflows within the compute organization.

A Bachelor's or Master's degree in Computer Science, Electrical Engineering, or a related field, coupled with at least 8 years of relevant industry or equivalent academic experience is essential. Proficiency in C and C++ programming, experience with large codebases, and a solid understanding of operating system interfaces for threads, process control, and virtual memory are required. You should also have experience writing and debugging multithreaded programs, a strong grasp of system-level architecture (interconnects, memory hierarchy, interrupts, memory-mapped IO), and a keen interest in system software. Excellent written communication skills are also necessary.