Fellow, ROCm GPU Libraries — AI/HPC Architect

AMD; Oregon; On-site; USD 180,000 - 260,000; Full time

Job summary

AMD is seeking a Fellow to lead the ROCm GPU libraries team, shaping AI and HPC performance libraries. You will direct vision, architecture, and design, mentor top talent, and drive strategic initiatives to push the limits of GPU software.

You will collaborate across hardware and software groups to align silicon features with software capabilities, engaging customers and partners to architect scalable solutions for demanding workloads.

Qualifications

• Deep expertise in parallel computing, scientific computing, and GPU software development.
• Advanced knowledge of GPU microarchitecture and programming models.
• Strong analytical and performance analysis skills.

Responsibilities

• Set the technical vision and roadmap for AI and HPC performance libraries for GPUs.
• Lead profiling and bottleneck analysis to maximize performance on current hardware.
• Engage with customers to understand workload requirements and architect high-performance solutions.
• Collaborate across hardware and software teams to influence future silicon and software features.
• Mentor engineers and leaders to develop the next generation of technical leaders.

Skills

GPU programming

C/C++

Performance optimization

Parallel computing

Leadership

Education

PhD or MSc in Computer Engineering/CS/Applied Mathematics

Job description

AMD is seeking a Fellow to lead the ROCm GPU libraries team, shaping AI and HPC performance libraries. You will direct vision, architecture, and design, mentor top talent, and drive strategic initiatives to push the limits of GPU software.

You will collaborate across hardware and software groups to align silicon features with software capabilities, engaging customers and partners to architect scalable solutions for demanding workloads.