Duyi Wang

Duyi Wang is a GPU AI Software Engineer at AMD, specializing in artificial intelligence and large language models (LLMs). Previously, Duyi was a core developer of xFasterTransformer, a high-performance inference framework for large language models.

Posts by Duyi Wang

Practical, Fault‑Robust Distributed Inference for DeepSeek on AMD MI300X

Learn how a small-radius expert parallel design with prefill–decode disaggregation enables scalable, fault-isolated LLM inference on AMD Instinct™ MI300X clusters.

November 12, 2025 by Peng Sun, Andy Luo, Gilbert Lei, Lingpeng Jin, Carlus Huang, Duyi Wang, Mingzhi Liu, Di Tian, Bill He, Jun Chen, Yutong Wu, Jiahao Zhou, Niko Ma