GPU Runtime & System Software Architect — Hybrid (San Jose)

Advanced Micro Devices; San Jose, Northern (CA, KY); Hybrid; USD 190,000 - 230,000; Full time

Job summary

Advanced Micro Devices in San Jose, CA, seeks a Fellow to design, develop, and optimize GPU runtime and system software for AI workloads and HPC deployments. You will evaluate new technologies, drive innovation, and collaborate across AMD teams and with customers from early product concepts to production.

The ideal candidate has deep expertise in GPU software, kernel and driver development in C/C++, concurrent programming, and a track record of leadership in shifting state-of-the-art

Qualifications

• Strong background and hands-on experience in GPU runtime and system software from low-level device drivers and OS internals up through applications and AI/ML frameworks.
• Deep expertise with kernel, driver and system level development in C/C++ and GPU software development and parallel programming.
• Deep expertise with modern concurrent programming and threading APIs.
• Proven history of communication software thought leadership, backed with patents, publications, and participation in industry standards bodies.

Responsibilities

• Provide deep technical leadership and guidance for GPU runtime and system features and technologies, optimizing across the entire AI software stack.
• Collaborate with hardware and software architects from early architecture concept to product deployment.
• Partner with top customers and hyperscalers to understand their workload requirements and deliver tailored architectural wins and software optimizations.
• Engage with executives and key stakeholders to provide insight into industry trends and recommend strategic initiatives.

Job description

Advanced Micro Devices in San Jose, CA, seeks a Fellow to design, develop, and optimize GPU runtime and system software for AI workloads and HPC deployments. You will evaluate new technologies, drive innovation, and collaborate across AMD teams and with customers from early product concepts to production.

The ideal candidate has deep expertise in GPU software, kernel and driver development in C/C++, concurrent programming, and a track record of leadership in shifting state-of-the-art