Student & Developer. Building cool things and exploring the world of tech.
I am a dedicated professional with a strong background in Machine Learning and Software Development. I am currently pursuing an MEng in Computer Science at Cornell Tech (expected 2026) and hold a BS in Computer Science from National Taiwan University (GPA 4.06/4.3).
I thrive on solving complex problems and creating value through innovative solutions. With a focus on continuous learning and improvement, I stay up-to-date with the latest industry trends and technologies.
Collaborated with a team of 8 to research balancing memorization and generalization in Large Language Models (LLMs). Designed and executed experiments to investigate the relationship, and visualized data to interpret results.
Taught AP computer science to 100 high school students. Led small group discussions and live coding demonstrations to enhance programming skills and understanding of advanced concepts.
Feb 2025 – June 2025
Analyzed and implemented a hybrid information retrieval system by fusing a lexical model (BM25) with a dense, BERT-based model (ANCE) to enhance search performance. Demonstrated that ML-based methods achieved superior in-domain performance on the FiQA dataset, while simpler, training-free methods proved more robust for cross-domain generalization on the TREC-COVID dataset.
View ProjectFeb 2024 – June 2024
Designed and developed a heuristic algorithm to solve an NP-hard vehicle scheduling problem, delivering high-quality, near-optimal solutions for large-scale instances in a fraction of the time required by exact MILP models.
View ProjectFeb 2024 – June 2024
Collaborated with NTU Hospital surgeons to analyze clinical data from over 3,000 liver cancer patients using random survival forests to predict postoperative recurrence and support medical decision-making.
View ProjectSep 2023 – Dec 2023
Developed a multilingual fake news detection system by fine-tuning and evaluating BERT-based models on diverse English and Japanese datasets to classify textual authenticity.
View ProjectFeb 2023 – June 2023
Predicted track danceability using six machine learning approaches, including regression models and neural networks. Used cross-validation and Adaboost to address overfitting.
View ProjectFeb 2025 - May 2025
Engineered and automated a cross-exchange arbitrage strategy. Achieved 50% ROI per month over two months with $2,000 capital, validating strategy robustness under live market conditions.
View ProjectSep 2023 - Dec 2023
Implemented a BERT base model and developed a tokenization algorithm for paragraph and span selection tasks. Won 2nd place out of 200 contestants in the Applied Deep Learning Course Competition.
View Project