Projects

BeTNet, a hybrid BERT-CNN classifier that detects phishing URLs at 98.33% accuracy and 99.31% sensitivity.

Decentralized uptime monitoring that makes SLA verification tamper-proof — validators stake, reach consensus on incidents, and are slashed for dishonesty.

AI e-commerce app that scores listing authenticity across Shopee, Lazada, and Amazon.sg from one search.

★ DSTA BrainHack 2025 Finalist

MERN social reading platform with AI review summaries, real-time book clubs, and playlist-style recommendations.

Full-stack profiling and optimization of GPU workloads — my final-year project, turning measurements into concrete speedups.

Java CLI hospital system with role-based access for four user types, appointment scheduling, and CSV-backed persistence, built on clean OOP design.

Deep-learning fraud detector reaching 94.29% precision on highly imbalanced data, with NLP-encoded metadata and LIME explainability.

X-to-Earn dApp on VeChainThor that rewards verified sustainable-fashion actions with redeemable B3TR tokens.

★ Top 6 · Easy x VeChain Singapore Hackathon

TensorFlow models that forecast port downtime and freight delays from historical global port data, served through a live dashboard.

Regression study predicting global game sales (R² > 0.7), comparing Linear, Lasso, and RBFN across a 1985–2016 chronological split.