Portfolio
Projects
A focused selection of product and engineering work across full-stack web development, Flutter mobile development and applied AI. Source code, live deployments and technical details are included where publicly available.
Syllabus Sync: Full-Stack Student Platform with Integrated AI Assistant
A full-stack student productivity platform for Macquarie University students, covering academic planning, deadlines and campus information. Sylla, an integrated AI assistant built into the platform, adds AI-assisted explanations, summaries, flashcards, quizzes and study planning. Selected for the Macquarie University Incubator.
Sylla – AI-Powered Study Assistant
An AI-powered study assistant with streaming responses, persistent conversation history and reusable study workflows for summaries, explanations, flashcards, quizzes and study planning. Designed to operate independently and as an integrated assistant within the wider Syllabus Sync ecosystem.
MQ Navigation – Flutter Campus Navigation App
Building a Flutter campus navigation app supporting destination discovery, route previews, transport information and access to key university locations. Implemented destination-based deep linking between Syllabus Sync and the mobile app, enabling direct web-to-mobile navigation, with a mobile-first, accessibility-focused interface and automated Flutter testing.
Astronomy Open Night – Flutter Event Navigation App
Co-developing a Flutter event navigation app for Macquarie University's Astronomy Open Night, supporting event information, venue navigation and in-app Google Maps walking directions. Currently in external testing and preparing for App Store and Google Play distribution. Developed by student app developers, not an official Macquarie University product.
CIFAR-10 Image Classification: PyTorch Benchmark
An end-to-end image-classification project comparing a custom CNN, MobileNetV2 and ResNet-18 under controlled training and transfer-learning conditions. ResNet-18 achieved 87.48% test accuracy. The project includes Grad-CAM interpretability, INT8 quantisation, CLI inference and a live Gradio deployment.