Advanced Micro Devices logo

Lead GPU Machine Learning & HPC Architect

Advanced Micro Devices
Full-time
On-site
Folsom, California, United States
Level - Senior

Position Overview

As a Lead GPU Machine Learning & HPC Architect at Advanced Micro Devices, you will play a vital role in the Radeon Technologies Group's efforts to develop cutting-edge GPU solutions designed for data centers and super-computers. The position is based in Folsom, CA, and involves close collaboration with a talented architecture and design team.

Job Overview

This role requires expertise in architecture exploration, modeling, and analysis of machine learning and high-performance computing workloads. You will evaluate new and emerging hardware and software technologies while contributing to the design and development of the next generation of AMD products.

Experience Level

We are seeking an individual with strong analytical and problem-solving skills, who is a capable team player with a focus on collaboration and mentorship. Strong communication, time management, and presentation skills are essential.

Key Duties

  • Communicate and collaborate with a network of architects and designers globally.
  • Identify complex technical challenges and propose potential solutions.
  • Work with architects to create innovative hardware solutions validated through various models and simulations.
  • Compile and summarize data or simulation results for architects and design teams.

Desired Qualifications

The ideal candidate will have a solid understanding of GPU architectures, a basic knowledge of CPU architecture, and experience with Network-on-Chip (NoC) design. Familiarity with machine learning frameworks such as TensorFlow and PyTorch, as well as Graphics and Compute APIs including CUDA, OpenCL, and Vulkan, is beneficial. Proficiency in programming languages such as C, C++, and Python, along with experience in hardware modeling and design using RTL or SystemC is preferred.

Education Requirements

A Bachelor's degree in Electrical Engineering, Computer Architecture, or Computer Science is required, with a Master’s or PhD degree preferred, along with relevant experience.