Master of Information Technology graduate (UNSW Sydney, 2026), focused on LLM applications, recommender systems and full-stack development.
Languages: Python, Java, JavaScript, Rust, SQL
AI / ML: LLM prompt engineering, multi-agent systems (LangGraph), recommender systems, PyTorch, scikit-learn
Apps: WeChat Mini Programs and Cloud Functions, HTML / CSS / jQuery, Java Swing, Android
Databases: SQL Server, relational modelling
Tools: Git, Unreal Engine 5, Figma, Pixso
| Project | What I did | Stack |
|---|---|---|
| MediChain: multi-agent LLM system for clinical diagnostic reasoning | Designed all six agent prompts, including the structured Critic ⇄ Diagnostician debate. Top-1 accuracy 100% on 7 ground-truth cases | LangGraph, FastAPI, RAG, MCP, React |
| Fashion Recommender: hybrid recommender on 8.5M Amazon reviews | Built the knowledge-based module for brand-new users and cut its batch runtime from 13.2 hours to about 15 seconds | Python, pandas, scikit-learn |
| Area | Repository | Projects |
|---|---|---|
| AI & machine learning | ai-ml-projects | MediChain, fashion recommender, fashion image classification, aerial scene classification |
| Backend & databases | backend-database-projects | RSheet concurrent spreadsheet engine (Rust), a SQL Server sales database |
| Full-stack & front-end | fullstack-frontend-projects | Two WeChat Mini Programs (one with a serverless cloud backend), Java hotel management system, Android app, a tea shop website |
| Games & interactive | game-interactive-projects | Playable UE5 underground station level |
| Product prototypes | product-prototype-projects | Accessibility-first consumer legal aid platform |
· Email: dannieec@hotmail.com · LinkedIn: LinkedIn

