tonycwliu

Meta GPU Techniques Team

                     Next Generation GPU Enablement (Blackwell, MI350)
                     High-performance Custom GPU Kernels
                     Advanced Algortihms for Large Language Models
                     Compiler & Ecosystem Advancements (Triton, GEMM Tuning)

Our work was recognized over $100 million infra annual saving, top-tier publications, & open-source contributions.

• Efficient Speculative Decoding for Llama at Scale: Challenges and Solutions, Arxiv, Aug 11, 2025

tonycwliu

#!/bin/sh
# This script help to update your Octopress blog by a single commend
# USAGE: sh blog.sh {your comment of this commitment}

# If no comment for this updating, simply say `Updating'
if [ $# -lt 1 ]; then
   commit_note=\'Updating.\'
else
   commit_note=\'$1\'
fi