Fourier Shuffle for Semantic Vision
A frequency–spatial upsampling method for medical image segmentation that improves validation performance while reducing memory use.
+1.77% validation · −13.8% GPU memoryOpen to research collaborationBrisbane, AU · 27.47° S / 153.03° E
Researcher · Engineer · Educator
Building models.Studying what’s inside.
+1.77%validation performanceagainst a U-Net baseline
−13.8%GPU memorywith the Fourier Shuffle model
$30K+sponsorship securedfor UQ Computing Society
1.5K+extension downloadsfor TraceForest
Selected work
Research that moves between theory and systems.
From frequency-domain vision to explainable code provenance, I am interested in methods that expose useful structure—and in building them far enough to test.
BlackBox Labs — TraceForest
An explainable detector for AI-generated code, combining stylometric AST features, random forests, SHAP, and developer tooling.
94% accuracy · 1.5K+ extension downloadsVision Transformer Attention Analysis
An experimental pipeline for extracting, visualising, and modifying internal attention distributions in pretrained vision transformers.
FLAN-T5 for Biomedical Lay Summarisation
A comparison of full fine-tuning, LoRA, and evolution strategies for translating specialist biomedical reports into accessible summaries.
>70% ROUGE · 2–3% trainable parameters with LoRA
About
More than a model and a metric.
I am a Machine Learning Researcher and Master of Data Science student atThe University of Queensland. My work focuses on computer vision, deep learning, and the mathematical structure behind model behaviour.
Alongside research, I teach subjects ranging from machine learning to theory of computing, lead industry partnerships for UQ Computing Society, and build tools designed to survive outside a notebook.
Across the stack
Research, teaching, leadership, and execution.
The common thread is turning difficult ideas into something other people can use, understand, or build upon.
Mar 2026 — Present
Research
Research Assistant
The University of Queensland
Deriving a Generalised Kaleidoscope transform for frequency-based autoencoders and supporting research on frequency-domain methods for computer vision.
Jul 2025 — Present
Teaching
Casual Academic
The University of Queensland
Teaching across machine learning, functional programming, theory of computing, and introductory software engineering.
Nov 2025 — Present
Leadership
Industry Officer
UQ Computing Society
Leading sponsorship outreach, pitching, negotiation, and partner relationships for one of UQ’s largest technology societies.
Jan 2026 — Mar 2026
Research
Summer Research Scholar
The University of Queensland
Designed a PsychoNet-based decoder and derived Fourier Shuffle for frequency–spatial upsampling in medical image segmentation.
Publication
Learning Frequency Domain Codes for Semantic Vision
Research on representations and operations that organise visual information in the frequency domain.
Under revision · OpenReview preprint
Read publication note →Latest talk
UQ Research Experiences Showcase
A presentation on frequency-domain image segmentation research completed through UQ’s Summer Research Program.
27 March 2026
Watch the talk →Let’s work together
Have an unusual machine-learning problem?
I am interested in research collaborations, technical projects, and conversations about frequency-domain vision, model behaviour, or tools that connect research with practice.