PhD AI Systems & GPU Performance Intern

AMD; San Jose (CA); Hybrid; USD 34,000 - 55,000; Full time

Job summary

AMD is seeking a PhD AI Systems & GPU Performance Engineering Intern in San Jose, CA. You will optimize AI models and training workflows on AMD Instinct GPUs, profiling with ROCm, Triton, and related tools.

You’ll develop benchmarking frameworks, analyze end‑to‑end AI workflows including LLMs, and deliver data‑driven recommendations to improve efficiency and throughput. This internship offers hands‑on work with cutting‑edge hardware and collaboration with researchers.

Qualifications

• PhD candidate with expected graduation after internship.
• Hands-on programming in Python and C/C++.
• Experience with Linux, profiling and performance analysis.

Responsibilities

• Profile, benchmark, and optimize AI training and inference workloads on AMD Instinct GPUs.
• Develop reproducible benchmarking frameworks and automation scripts.
• Analyze end-to-end AI workflows including LLMs and generative AI.
• Provide data-driven recommendations to improve efficiency, throughput, and latency.
• Collaborate with engineers and researchers on performance investigations.

Skills

Python

C/C++

Linux scripting

Profiling

Performance analysis

Education

PhD in Computer Science/Engineering/Applied Mathematics

Tools

CUDA

HIP

Triton

ROCm Profiler

ROCm

Job description

AMD is seeking a PhD AI Systems & GPU Performance Engineering Intern in San Jose, CA. You will optimize AI models and training workflows on AMD Instinct GPUs, profiling with ROCm, Triton, and related tools.

You’ll develop benchmarking frameworks, analyze end‑to‑end AI workflows including LLMs, and deliver data‑driven recommendations to improve efficiency and throughput. This internship offers hands‑on work with cutting‑edge hardware and collaboration with researchers.