Senior GPU/ML Compiler Architect — Performance Lead

AMD; San Diego (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

AMD is seeking a strategic software engineering lead to advance performance of applications and benchmarks on future and recent GPU architectures. You will contribute to language implementation and optimization within the open source LLVM framework and collaborate with cross-functional teams on cutting-edge compiler technology.

The role emphasizes leadership, high-quality code, and strong communication as you work across AMD to improve compiler technologies for AMD GPUs and associated software

Qualifications

• Strong object-oriented programming background in C/C++ (or C++).
• Experience with modern concurrent programming and threading APIs.
• Ability to lead complex issues to resolution and communicate effectively with cross-functional teams.

Responsibilities

• Improve performance of key applications and benchmarks for future and recent AMD GPU architectures.
• Design and develop groundbreaking AMD technologies in the LLVM compiler framework.
• Collaborate with architecture teams to guide optimization efforts and bring up new ASIC/hardware.
• Debug, investigate, and propose efficient alternatives for performance bottlenecks.
• Build technical relationships with peers and partners and contribute to open source LLVM projects.
• Work with ML frameworks and HPC apps to identify compiler enhancements for AMD GPUs.

Skills

C/C++

Object oriented programming

Parallel/concurrent programming

Leadership

Communication

Education

Bachelor's degree in Computer Science/Engineering

Tools

GitHub

LLVM

Profilers

Debugging tools

Job description

AMD is seeking a strategic software engineering lead to advance performance of applications and benchmarks on future and recent GPU architectures. You will contribute to language implementation and optimization within the open source LLVM framework and collaborate with cross-functional teams on cutting-edge compiler technology.

The role emphasizes leadership, high-quality code, and strong communication as you work across AMD to improve compiler technologies for AMD GPUs and associated software