Shaun is an electrical/biomedical engineer based in Sydney, Australia.
What's he interested in?Developing technology which improves human wellbeing.
His recent academic research focuses on addressing limitations with deep learning methods applied to medical imaging modalities such as electrocardiography (ECG).
Why trust him?His technical skills have been developed across a broad range of industry roles:
- Electronics product development for implantable medical devices
- Manufacturing process development for a cell therapeutics scale-out
- Simulation software and systems engineering for electric boats
Contact details at bottom of page.
Awards
- 2025 Weights & Biases LLM Evals Competition - 1st Place
- 2024 Peter Farrell Cup - 1st Place Trailblazer
- 2023 VEX University Robotics World Championship - Design Award
- 2022 VEX University Robotics World Championship - Division Finalist
- 2021 University of Auckland Engineering Honours Thesis - Best Project (Electronic Systems category)
Evaluating reasoning ability of frontier Large Language Models such as Claude, Gemini, ChatGPT.
Prototyped system which would enable continuous ECG monitoring inside of moving vehicles (i.e. ambulances) for patients with evolving acute heart conditions (e.g. heart attack).
Publications
2025 Adapting DNN models to athletic ECGAthlete hearts confuse automated ECG interpretation programs, leading to misdiagnoses and a need for clinicians specializing in sports cardiology.
My masters thesis conducted at the University of New South Wales looks at how state-of-the-art ECG interpretation programs based on deep learning models can be modified for this unique patient cohort.
A unique challenge is the small volume of publically available athletic ECG datasets. I explore the use of domain adaptation methods to analyze and solve this problem.
Result: My methods improve the diagnostic performance of a reference model developed by Seoul National University researchers on both athletic and general populations.
2021 Evaluating Common Electronic Components and GaN HEMTs Under Cryogenic Conditions
[paper]