Principal System Software Engineer - CUDA Driver

NVIDIA•Us, CA

About The Position

NVIDIA is hiring a senior engineer to work on the CUDA driver, a core component of its platform for accelerating general-purpose computation on GPUs. The CUDA driver helps deliver features and improvements that realize the potential of NVIDIA hardware across workloads including deep learning, scientific computation, self-driving cars, video games, and virtual reality. It interacts with GPU hardware, kernel-mode drivers, switches, and the operating system to provide a unified programming model across system configurations and hardware capabilities. As a member of the team, you will use design abilities, coding expertise, and creativity to develop the CUDA compute platform. You will craft solutions to challenging problems and help shape CUDA’s future direction while collaborating with peers across NVIDIA. The role includes evangelizing, architecting, and implementing CUDA features; driving development across teams; collaborating with hardware architecture teams; improving CUDA APIs and the programming model; designing and maintaining performance and precision modeling; writing effective, maintainable, well-tested code; and developing for multiple operating systems.

Requirements

Bachelor of Science or Master of Science degree in Computer Science, Electrical Engineering, or a related field, or equivalent experience.
15+ years of relevant systems software development experience.
Strong C programming skills.
Experience designing, debugging, and maintaining complex software stacks.
Background with operating system interfaces for threads, process control, and virtual memory.
Experience with hardware/software co-design, performance modeling using emulation or simulation, and creating software programming-model exposures for hardware features.
Expert interpersonal, verbal, and written communication skills, with the ability to achieve objectives under tight deadlines.
Strong collaborative and interpersonal skills, including a proven ability to guide and influence effectively in a dynamic matrix environment.

Nice To Haves

Understanding of system-level architecture, such as interconnects, memory hierarchy, interrupts, and memory-mapped I/O.
Experience designing and implementing drivers for rich hardware acceleration engines and software verification test plans.
Knowledge of CPU and GPU architectures, and memory coherence and consistency models.
Some background with kernel-mode development.
Some familiarity with C++.

Responsibilities

Evangelize, architect, and implement new CUDA features.
Oversee and drive development efforts across multiple teams.
Collaborate with members of hardware architecture teams.
Help define forward-looking improvements to the CUDA APIs and programming model.
Design and maintain performance and precision modeling.
Develop effective, maintainable, and well-tested code.
Develop code for multiple operating systems.