|
Computer Engineering student at Temasek Polytechnic — building smart systems with AI/ML, IoT, and full-stack development.
I'm a Computer Engineering student at Temasek Polytechnic (2023–2026), passionate about building technology that solves real problems. I love working at the intersection of software, hardware, and intelligent systems.
From designing AI-powered self-checkout systems using Raspberry Pi and YOLO, to crafting mobile finance apps and live operations dashboards — I thrive on turning complex challenges into clean, practical solutions.
Currently working part-time as an IT Analyst at Green Garden Integrated Services, building dashboards and AI agent prototypes to streamline daily operations.
A deep learning web app that classifies musical instruments from uploaded images in real time. Implements a CNN with ResNet50 transfer learning for improved accuracy. Uses data augmentation, dropout regularization, and early stopping to prevent overfitting. A Flask API serves the model, allowing users to upload images and receive instant predictions.
A desktop utility billing management application built with Java OOP principles. Features include customer account management, bill generation, payment tracking, and usage reports — demonstrating core concepts of inheritance, encapsulation, and polymorphism.
A portable self-checkout prototype using 3 Raspberry Pis. RPi 1 handles barcode scanning and blocked-item logic via Pi Camera. RPi 2 runs YOLO object counting via overhead camera. Alerts are sent via MQTT (Mosquitto) when scanned and detected item counts don't match — displayed on a customer kiosk and admin web panel.
Cross-platform mobile app in Flutter with income/expense tracking, interactive charts, budget tools, savings goals, Google login, profile image upload/crop, social sharing, daily reminders, and recurring transaction support.
Home monitoring system built with Raspberry Pi and a breadboard sensor circuit tracking temperature, humidity, and pressure in real time. Sensor readings are published via MQTT (Mosquitto) and received live on a mobile app using EasyMQTT.
Cleaned, analyzed, and visualized 50 years of global storm data using KNIME for ETL and data modelling, and Tableau for interactive dashboards detecting storm type patterns worldwide.
Click any project to explore details, images, and tech used.
Completed in the following modules:
Open to internships, collaborations, and interesting projects. Feel free to reach out anytime!