About
I'm Matheesha Ileperuma, a final-year Computer Science (AI) student at Swinburne University of Technology, Melbourne. I design and build AI systems that actually ship — offline LLMs for remote communities, production-grade recommendation engines with full CI/CD pipelines, and ML classifiers deployed behind real APIs.
My work spans RAG pipelines, fine-tuned language models, full-stack web applications, and everything in between. When I'm not training models, I'm chasing down a JDM part or digging through a record crate.
Currently looking for grad roles and internships in AI/ML engineering — Melbourne-based or remote.
Selected Work
01 / 06 — FEATURED
Active · Fully DeployedAI-powered cross-media recommendation engine. Learns your taste through a conversational onboarding chatbot, then surfaces one personalised pick per media type — books, films, articles — with LLM-written explanations and cross-media linking. Built with a full CI/CD pipeline blocking pushes to main.
Architecture
Vite + Tailwind frontend on Vercel → FastAPI + SQLModel backend on Render → PostgreSQL 16 with pgvector for semantic search → catalogue adapters: Google Books, TMDb, Guardian, YouTube → JWT auth, offline demo mode, returning-user memory via LLM context injection.
Status: 5 Sprints Complete
Onboarding chatbot · Recommendation engine · Cross-media linking · Feedback system (love / dislike / save / skip) · Full CI pipeline with lint, tests, and build gates blocking main.
02 / 06
Capstone · UniversityOffline LLM chatbot preserving traditional ecological knowledge for an indigenous community in Bangladesh. Runs entirely on local hardware — zero internet dependency. Led RAG pipeline design and LLM evaluation (Mistral 7B vs Phi-4 14B); wrote sprint documentation and ran client demos.
03 / 06
Content-based product recommender across 125 products in 14 categories. Builds a TF-IDF matrix from combined product features, computes cosine similarity, and explains each suggestion in plain English. No framework — pure JS frontend.
GitHub →04 / 06
Upload a CSV of bank transactions → ML classifies spending categories → interactive Recharts dashboard. User corrections feed back into retraining. Trained on 1,600 labelled transactions, 100% test accuracy.
05 / 06
AAPL forecasting with LSTM, GRU, SimpleRNN, and an ARIMA ensemble built across 7 checkpoints of increasing complexity. Best model: Bidirectional GRU (2 layers, 128 units). Baseline LSTM R² = 0.965. Includes 5-day multistep forecasting.
06 / 06
ML applied to cybersecurity: spam detection (TF-IDF + Logistic Regression / Naive Bayes / Linear SVM) and malware detection (PE header features + Random Forest). Full EDA reports with word clouds, class distributions, and cross-model comparisons.
Capabilities
AI / LLM
Machine Learning
Backend
Frontend
Tools & Infra
Get in Touch
Open to grad roles, internships, and interesting problems — Melbourne or remote.