AI & Data
Product Engineer

Research Paper

Developed a context-aware AI system fusing YOLOv11 and MediaPipe to identify threats based on human pose in partnership with AiLECS Lab. Integrated Vision-Language Models (VLMs) to provide natural language explanations, moving beyond simple bounding boxes to contextual scene understanding.

Full-Stack Application

Built a full-stack desktop application as part of a student team to help Teaching Assistants equitably grade group coding projects. The tool automates the extraction of Git metrics (commits, LoC, churn) to visualise student performance and generates scalable grading sheets for Moodle. It also features an AI-assisted summary mode (Gemini) to provide qualitative context on student work patterns

Creative Coding

A macOS screensaver that updates daily and transforms live global and Australian headlines into unique 3D movie posters. The system aggregates breaking news (BBC, CNN, ABC) and utilises Nano Banana to generate cohesive "Pixar-style" claymation visuals. It leverages Google Gemini 2.5 with real-time search grounding to ensure the accompanying text summaries are factual and up-to-date.

Learning Project

Conducted a comparative study on a dataset of 2,000 websites to classify malicious phishing attempts using various AI and Machine Learning algorithms. The project evaluated Neural Networks (ANN), SVMs, and Ensemble methods to determine optimal classification strategies. A parameter-tuned Random Forest model emerged as the most robust solution on this specific dataset.