Senior GPU ML Training Performance Engineer

Advanced Micro Devices; San Jose (CA); On-site; USD 130,000 - 170,000; Full time

Benefits offered by this job

Comprehensive health benefits

Inclusive workplace culture

Opportunities for career advancement

Job summary

A leading semiconductor company is looking for a Principal/Senior GPU Software Performance Engineer in San Jose, CA. The role involves optimizing post-training workloads on AMD Instinct GPUs, improving throughput, and collaborating with various teams to drive measurable performance improvements. The ideal candidate should possess proven GPU performance engineering experience, strong skills in Python and PyTorch, and hold an advanced degree in a relevant field. Benefits and a supportive work culture are part of the offer.

Qualifications

• Proven experience in GPU performance engineering for deep learning.
• Comfortable writing reading kernels in C++.
• Strong background in conducting performance optimizations.

Responsibilities

• Lead performance optimization for finetuning and RL training on AMD GPUs.
• Collaborate to improve throughput and memory efficiency.
• Optimize multi-GPU training patterns and diagnose performance issues.

Skills

GPU performance engineering

Proficient in Python

Strong PyTorch experience

Hands-on with SFT, LoRA

Experience with distributed systems

Education

B.S./M.S./Ph.D. in Computer Science or related field

Tools

ROCm/HIP

Triton

Job description

A leading semiconductor company is looking for a Principal/Senior GPU Software Performance Engineer in San Jose, CA. The role involves optimizing post-training workloads on AMD Instinct GPUs, improving throughput, and collaborating with various teams to drive measurable performance improvements. The ideal candidate should possess proven GPU performance engineering experience, strong skills in Python and PyTorch, and hold an advanced degree in a relevant field. Benefits and a supportive work culture are part of the offer.