Compute System Architect

The Compute System Architect team is dedicated to GPU compute and memory subsystems. As a member of this team, you will explore sophisticated cross-unit features in modern GPU architectures.

Your responsibilities will include:

• Exploring innovative GPU composition and novel system functionalities related to processing and storage.
• Connecting with GPU architecture designers to create advised and random functional testing plans.
• Developing and improving infrastructure and methodology to generate tests.
• Generating, running, and debugging tests in various platforms, such as functional simulators, full-chip and unit-level RTL, emulators, and silicon.
• Building innovative tools to improve efficiency.

To succeed in this role, you will need:

• Knowledge of computer architecture, compiler, and assembly language.
• At least 4 years of working experience.
• A good understanding of GPU concepts and pipelines, including parallel computing and memory systems.
• Familiarity with Linux.
• Proficiency in C++ and Python development.
• A bachelor's degree in Computer Science or Electrical Engineering; a master's or Ph.D. or equivalent experience is a plus.

To stand out from the crowd, consider highlighting:

• Knowledge and experience with CUDA programming and debugging.
• Random system development and work experience.
• ASIC experience.
• Compiler experience.