Work.

Co-founded a mobile app to help people discover and connect with others. Full product ownership from architecture through implementation as Dev Lead.

Co-founder · Full product ownership · MVP ETA Jan 2027

Built an internal tool to generate validated real South Australian addresses based on socio-economic status (SEIFA indices) and remoteness classifications — filling a gap no public API could address. All outputs verified via Mapbox API.

Validated address generation · SEIFA + remoteness filtering · Internal QA tool

Scraped the UC Santa Barbara American Presidency Project to collect ~180 presidential debate transcripts and ~25,000 campaign documents, with multi-threaded processing, rate limiting, and full metadata extraction.

~25,000 documents collected · Decades of US political communication data

Built the first intelligence analytics capability within the CBS Prevention Team — integrating ABS, SA Health, ACCC, and DataSA data into unified dashboards and GIS maps used by the Minister's Office.

Minister's Office reporting · SOPs institutionalised

A clinician-facing and patient-facing mental health mobile app developed for the University of Melbourne's Psychiatry research group. Migrated from Uniapp to Expo React Native, reducing costs by ~$500/month and delivering a GDPR-compliant production application.

~$500/month cost saved · Production deployed · Cross-platform iOS + Android

Contributed to operational intelligence and crime analytics as an Intelligence and Coordination Officer. Built dashboards and analytical products supporting frontline policing, resource allocation, and command-level decision-making across South Australia.

Operational decision support · Crime pattern analytics · Cross-unit coordination

Automated flow cytometry data analysis using cloud and HPC, developed test infrastructure for improved reproducibility, and contributed to the open-source celseq2 workflow toolkit.

Reduced manual processing · celseq2 open-source contribution

Designed a two-stage automated fact-checking system for climate change claims — TF-IDF evidence retrieval followed by Transformer-based classification (SUPPORTS / REFUTES / NOT ENOUGH INFO / DISPUTED), outperforming LSTM baselines.

Transformer model outperformed LSTM · Scalable misinformation detection pipeline

Built and refined interactive digital educational content aimed at bridging the gap between education and professional success. Conducted market research on leveraging advanced technologies to improve course engagement and learner outcomes for students across Australia and beyond.

80+ students supported · Improved course engagement through technology-driven content innovation

Research project exploring AI/ML techniques for autonomous database management — covering workload-driven optimisation and automatic index selection to reduce DBA overhead and improve query performance.

Literature review · Autonomous DB optimisation frameworks

Harvested and analysed Twitter and Mastodon data alongside ABS SUDO spatial data to produce a Social Sense Dashboard illuminating Australian sentiment, social trends, and regional behavioural patterns.

Cross-platform social analytics · Spatial + social data fusion

Processed a large-scale Twitter dataset on the University of Melbourne's SPARTAN HPC cluster using MPI and Python, identifying tweet distribution across Australian cities and top contributors.

HPC parallel processing · City-level tweet distribution insights

Built AutoRegressive time series models with rolling window forecasting to quantify how El Niño-Southern Oscillation patterns amplify commodity price volatility and food security-induced conflict risk.

Climate → conflict risk insights · 10-month CSIRO research engagement

Implemented AI agents for Cachex — a two-player connection game — covering heuristic A* search for the exploration phase and a competitive game agent using strategic anticipation and sabotage logic.

A* pathfinding · Competitive AI agent · Game theory application

Built a Python automation script to streamline HPLC experiment result processing, and applied unsupervised clustering methods to uncover hidden patterns in complex medical research datasets.

Reduced HPLC processing time · Improved data quality confidence

Collaborative full-stack CRM system built as the COMP30022 IT Project — managing customer interactions and data with a responsive React frontend, Express REST API, and MongoDB backend.

Full-stack delivery · Team project · Production-grade architecture

Analysed 2014–2017 New York City Yellow Taxi data using Apache Spark for big data processing, applying linear modelling to predict trip costs and clustering to identify traffic hotspots.

Big data Spark pipeline · Traffic hotspot clustering · Trip cost modelling

Five major iterations of a personal website — currently rebuilt in v5 with a Nothing OS-inspired minimal design language. A living record of technical and professional development since 2020.

5 major versions · rin.contact · Open source