Sean Song

Sean Song#

Sean is a Machine Learning Engineer at AMD, where he specializes in research and deployment of state-of-the-art deep learning solutions on AMD’s AI accelerators. With a PhD in Mathematics and Computational Science, he brings over a decade of rich research and development experience, with a focus on machine learning (ML) and high-performance computing (HPC). Sean’s current expertise extends across diverse domains that encompass computer vision, multi-modal learning (vision and language), and large language models (LLMs).

Posts by Sean Song

January 22, 2026

LLM Inference Optimization Using AMD GPU Partitioning

Demonstrate how to leverage compute and memory partitioning features in ROCm to scale model serving.

https://rocm.blogs.amd.com/software-tools-optimization/multi-inf-engine-gpu-partition/README.html

June 09, 2025

LLM Quantization with Quark on AMD GPUs: Accuracy and Performance Evaluation

Learn how to use Quark to apply FP8 quantization to LLMs on AMD GPUs, and evaluate accuracy and performance using vLLM and SGLang on AMD MI300X GPUs.

https://rocm.blogs.amd.com/artificial-intelligence/quark/README.html

February 09, 2025

PyTorch Fully Sharded Data Parallel (FSDP) on AMD GPUs with ROCm

This blog guides you through the process of using PyTorch FSDP to fine-tune LLMs efficiently on AMD GPUs.

https://rocm.blogs.amd.com/artificial-intelligence/fsdp-training-pytorch/README.html

January 24, 2025

Vision Mamba on AMD GPU with ROCm

This blog explores Vision Mamba (Vim), an innovative and efficient backbone for vision tasks and evaluate its performance on AMD GPUs with ROCm.

https://rocm.blogs.amd.com/artificial-intelligence/vision-mamba/README.html

November 01, 2024

Distributed Data Parallel Training on AMD GPU with ROCm

This blog demonstrates how to speed up the training of a ResNet model on the CIFAR-100 classification task using PyTorch DDP on AMD GPUs with ROCm.

https://rocm.blogs.amd.com/artificial-intelligence/ddp-training-pytorch/README.html

October 23, 2024

Inference with Llama 3.2 Vision LLMs on AMD GPUs Using ROCm

Meta's Llama 3.2 Vision models bring multimodal capabilities for vision-text tasks. This blog explores leveraging them on AMD GPUs with ROCm for efficient AI workflows.

https://rocm.blogs.amd.com/artificial-intelligence/llama3_2_vision/README.html