Kuan-Yi (Gary) Lee

I am currently a Master's student in ECE at Carnegie Mellon University (Taiwan GSSA Scholar), focusing on the systems side of machine learning — from efficient training and inference to deploying large speech and language models at scale.

I have 3 years of experience in ML and AI applications, having interned at MediaTek, NXP, and ASUS. My work has involved optimizing LLM training pipelines, developing speech-text aligned tokenizers, and building multi-agent systems for manufacturing and anomaly detection.

Beyond technical work, I enjoy teaching and mentoring in the ML space. I have served as a teaching assistant for multiple courses at NTU, including "Machine Learning" and "Deep Learning for Computer Vision," where I designed homework assignments and developed automated evaluation pipelines for hundreds of students. Additionally, I led a professional workshop on AI Agents and Multimodal LLMs, delivering lectures to over 320 professionals and students.

ML Systems LLM Inference Speech AI Distributed Training