Menghsuan Yang

Menghsuan Yang is a Senior Member of Technical Staff on AMD’s Data Center GPU team, focusing on AI operator kernel development and performance optimization. Menghsuan joined AMD in 2025 after working at MediaTek, where he contributed to the AI compiler team with a focus on operator-level tuning, model performance optimization, and multi-core execution efficiency. His areas of interest include AI compiler optimization, GPU kernel tuning, and efficient deep learning inference.

Posts by Menghsuan Yang

Avoiding LDS Bank Conflicts on AMD GPUs Using CK-Tile Framework

This blog shows how CK-Tile’s XOR-based swizzle optimizes shared memory access in GEMM kernels on AMD GPUs by eliminating LDS bank conflicts

July 25, 2025 by Haocong Wang, Clement Lin, Menghsuan Yang, Yu-Chen Lin, Bobo Fang, Chunhung Wang, David Li, George Wang, Anshul Gupta