Pin-Ying Wu

About Pin-Ying Wu

I am a machine learning engineer at TSMC, focusing on LLM-based solutions for data-driven business decisions that enhance decision-making efficiency. Specifically, I developed an interactive chatbot for internal pricing decisions, leveraging cutting-edge information retrieval and summarization techniques, which achieves 92.1% average recall on the internal data. Prior to that, I received my M.S. degree from the Department of ECE at University of California San Diego (UCSD). I am an AI enthusiast and my interests and domain knowledge span across Generative AI, Computer Vision, and Deep Learning. Aside from industrial experience, I had rich research experience at research labs. I was a research intern investigating the reliability of LLMs in the context of 3D Visual Question Answering (3D-VQA) in Statistical Visual Computing Lab at UCSD, advised by Prof. Nuno Vasconcelos. Before joining UCSD, I received my B.S. degree from the EE department at National Taiwan University (NTU), where I was fortunate to work with Prof. Yu-Chiang Frank Wang at Vision and Learning Lab on Audio-Visual Learning for Sounding Source Localization.

Experience

1. Machine Learning Engineer

   台積電

   Apr 2024 — Present · San Jose, CA, US

   Developing LLM-based solutions for data-driven business decisions that enhance decision-making efficiency- Leading an interactive chatbot project for internal pricing decisions, leveraging cutting-edge information retrieval and summarization techniques (such as RAG and CoT), which achieves 92.1% average recall on the internal data- Coordinating across teams to refine project requirements, address challenges, and ensure continuous alignment between technical deliverables and strategic business goals.

Education

• UC San Diego

  Master of Science - MS

  2022 — 2024

• National Taiwan University

  Bachelor of Science - BS

  2018 — 2022