Personal, academic, and open-source work.
An AI agent that fetches recent ArXiv research papers based on user search — like Google Scholar but in a more user-friendly and readable format. Built with LLMs and agentic retrieval pipelines.
Developed a machine learning model to accurately detect whether an essay was written by a student or an LLM. Finished in the top 25% of the Kaggle competition leaderboard.
Self-driving prototype using CNNs trained on dash-cam images to predict steering angles in real time. Built as part of the Team Camber Racing autonomous vehicle project.
Python package fetching real-time International Space Station data — location, crew, pass times and more. Over 60,000 downloads on PyPI.
Computer vision pipeline for road lane identification using Canny edge detection, Hough transforms, and perspective warping — part of the self-driving vehicle stack.
Applied logistic regression to a large imbalanced dataset, achieving 98.5% accuracy — outperforming ANN, random forest, decision tree, and SVM. Research presented at NUS.
Android app helping visually impaired users localize and identify objects within a camera frame using real-time image processing and voice feedback.