Open to opportunities — Malaysia
MSc in Computer Science, UNIMAS. I train convolutional neural networks for image classification — and ship Flutter apps, Laravel web apps and Python / Electron desktop tools.
I hold an MSc in Computer Science and a BSc in Cognitive Science (Hons) from Universiti Malaysia Sarawak. My research focuses on image classification with CNNs — benchmarking architectures like VGG, Inception and DenseNet, then pushing accuracy further with feature fusion and stacking ensembles, reaching ~94.9% on dog-breed classification.
Beyond research, I like shipping real software: a coffee-shop ordering app in Flutter, an appointment-booking system in Laravel, a multi-clip video editor in Electron, and an AI subtitle generator powered by faster-whisper.
Fuses deep features from multiple pre-trained CNNs with classical ML classifiers on the Stanford Dogs dataset — beating every single-model baseline.
Benchmarks AlexNet, VGG, GoogLeNet, InceptionV3, Inception-ResNet and DenseNet across learning rates and dataset sizes.
Stacks multiple CNN backbones with a meta-learner to push classification accuracy beyond single-model baselines.
Coffee-shop ordering app with the full flow: login → menu → product details → cart → checkout.
Colour-coded BMI calculator with instant category feedback.
Student-records app with live-syncing create / read / update / delete on Firebase Realtime Database.
Monitors Douyin live rooms with a foreground service and auto-records the stream the moment a room goes live.
Laravel 10 appointment booking & scheduling web app with a role-based admin dashboard.
Multi-page responsive static website built with Bootstrap.
Windows multi-clip video editor with timeline trim / split and FFmpeg-powered export to a single file.
Local CLI/GUI tool using faster-whisper to generate Chinese subtitles (.srt / .vtt / .ass and more) on CPU or CUDA.
Universiti Malaysia Sarawak (UNIMAS)
Research: dog-breed classification via feature fusion of pre-trained CNNs and ML classifiers — ~94.9% accuracy.
Universiti Malaysia Sarawak (UNIMAS)
Final Year Project: benchmarking CNN architectures for dog-breed recognition.
I'm currently open to roles in AI / machine learning and software development. The fastest way to reach me is LinkedIn.