Senior ML Compiler Engineer for AI Hardware Accelerators

AMD; San Jose (CA); On-site; USD 120,000 - 170,000; Full time

Job summary

AMD is seeking a motivated engineer to join a core team advancing a cutting-edge machine learning model compiler for AMD AIE in Ryzen AI PC and embedded FPGA devices.

You will integrate compilers with ML frameworks, implement partitioning, and develop runtimes to distribute work across x86 and AIE cores, contributing to innovative AMD technologies while collaborating with diverse teams.

Qualifications

• Strong object-oriented programming background, C/C++ and Python.
• ML Compiler and Runtime technologies such as MLIR, LLVM.
• Compiler building skills.
• Code generation for a ML hardware accelerator.
• Machine Learning concepts and model development experience.
• Understanding of ONNX, PyTorch, etc.
• Ability to write high quality code with a keen attention to detail.
• Experience with modern concurrent programming and threading APIs.
• Experience with software development processes and tools such as debuggers, source code control systems (GitHub) and profilers is a plus.
• Effective communication and problem-solving skills.

Responsibilities

• Integrate compiler and compiled model with ML Frameworks (such as ONNX and PyTorch)
• Implement model partitioning in ML Frameworks and/or MLIR
• Implement runtime to distribute work to and collect results from x86 cores and AIE cores
• Design and develop new groundbreaking AMD technologies
• Debug and fix existing issues and research alternative, more efficient ways to accomplish the same work

Skills

C/C++

Python

MLIR

LLVM

ONNX

PyTorch

ML Compiler

Code generation

Concurrency

GitHub

Education

Bachelor’s or Master’s in Computer Science/Engineering

Tools

GitHub

Debuggers

Profilers

Job description

AMD is seeking a motivated engineer to join a core team advancing a cutting-edge machine learning model compiler for AMD AIE in Ryzen AI PC and embedded FPGA devices.

You will integrate compilers with ML frameworks, implement partitioning, and develop runtimes to distribute work across x86 and AIE cores, contributing to innovative AMD technologies while collaborating with diverse teams.