Principal ML Performance Engineer
Advanced Micro Devices, Inc.
Optimize machine learning performance on AMD GPUs using ROCm, Python, and C++ while guiding sophisticated engineering issues toward resolution.
AMD seeks a Principal Software Development Engineer with strategic leadership and strong performance optimization skills. You will improve inference and training workloads across PyTorch, vLLM, llama.cpp, ROCm libraries, and GPU kernels. The work spans multiple software layers and open-source communities.
Key Responsibilities:
- Analyze machine learning models running on AMD GPUs
- Optimize Python and C++ across ROCm software layers
- Profile and benchmark workloads to target improvements
- Support hardware bring-ups and future product improvements
- Contribute to upstream open-source projects and technical relationships
Requirements:
- Strong programming background in C or C++ and Python
- Experience with PyTorch, vLLM, or llama.cpp
- GPU programming or performance optimization experience is beneficial
- Experience with Linux, concurrency, debuggers, GitHub, and profilers
- Bachelor’s or Master’s degree in a relevant engineering field
Lead AMD GPU optimization through ROCm, profiling, concurrent programming, and thoughtful evaluation of software performance.
#J-18808-LjbffrReference: WJ-4483_1629103 · Posted 15 Nov 2026, 02:03 AM