Rukshan Dias.
Data Scientist & Software Engineer Melbourne
I build data products and web apps on live open data, from forecasting models to full-stack platforms.
Featured Projects
Melbourne Pulse – Live CBD Pedestrian & Parking Forecasting
Personal project, live platform
- The problem
- Melbourne has no simple live view of how busy the CBD is right now, or how busy it will be over the next 24 hours.
- What I built
- An hourly GitHub Actions pipeline with a Pytest validation gate that ingests City of Melbourne open data into Neon Postgres, a LightGBM 24-hour forecast with no future leakage, and Gemini Flash plain-English summaries.
- What it showed
- The forecast beat the 8-week historical baseline by 10% on 132,696 unseen sensor-hours, with the whole platform running on free tiers.
- more accurate than baseline
- 10%
- test sensor-hours
- 132,696
- stations and sensors
- 99 + 6,300+
- running cost
- $0/month
MOP Use Case Quality & Publication
Team Lead, Melbourne Open Playground (Chameleon)
- The problem
- Data science use cases across a 200+ use case repository needed consistent quality before they could be published.
- What I built
- Jupyter Notebook standards, technical and GitHub reviews, progress tracking, and a smoother publication workflow with the Quality Management and Web teams, while mentoring Data Scientists.
- What it showed
- Use cases reached a publication-ready standard through a repeatable review process.
- use cases in the repository
- 200+
- commits, #1 contributor
- 64
- Team lead
Momentum Todo
Personal project, live app
- The problem
- Daily planning tools tend to be either too cluttered or too basic.
- What I built
- A production-ready productivity platform with task management, project organisation, calendar planning, progress tracking, authentication, cloud persistence and a responsive UI.
- What it showed
- A complete full-stack product, deployed and live on Vercel.
- Live
- Clerk auth
- Neon Postgres
Invoice Management System
Internal tool, built during my Irys software engineering internship
- The problem
- Invoices, clients and billing cycles were hard to track internally.
- What I built
- A secure, data-driven web app, translating the business requirements into database schemas and technical specifications.
- What it showed
- Improved internal tracking and operational efficiency.
- Flask
- Supabase
- Secure web app
More Projects
Weather Impact on Public Transport
Predicts Melbourne transport congestion from BOM rainfall data with ML, plus an LLM layer that explains results.
Melbourne Housing Price Prediction
A Streamlit house-price app built on 225 records, 4 models, 5-fold CV and SHAP, with a full PDF report.
Wheelchair Navigation System (AI)
Plans wheelchair-accessible routes with A* search, putting terrain and accessibility ahead of shortest distance.
Wumpus World AI Planning (PDDL)
Solves the Wumpus World in PDDL with sensing and constraint reasoning, comparing several planning solvers.
Experience & Education
Professional Experience
Data Science Team Lead
Chameleon - Smarter World Melbourne
Jul 2026 – Oct 2026
- Led Data Science delivery, building a centralised work allocation and tracking process that made ownership, workload and progress visible across the team.
- Built the first reference implementation of the team's technical standards, used to guide consistent delivery across future Data Science work.
- Led technical reviews across code, analytical notebooks, data visualisations and documentation, driving quality improvements before release.
- Mentored team members on Git/GitHub, pull request workflows, code quality and technical troubleshooting.
- Ranked as the #1 repository contributor with 64 commits while leading the team.
- Python
- Jupyter
- Pandas
- Data Visualisation
- Git
- GitHub
- Microsoft Planner
- Excel
Data Scientist / Data Science Project Contributor
Chameleon - Smarter World
Mar 2026 – Jun 2026
- Owned a transport analytics use case for the Melbourne Open Playground (MOP) project, examining how weather conditions affect public transport activity.
- Built and maintained data pipelines to clean, integrate and transform transport, weather and event datasets.
- Conducted exploratory analysis and data visualisation to uncover patterns and actionable insights.
- Managed development through Git/GitHub reviews and delivered the solution to MOP's READY TO PUBLISH repository.
- Python
- Pandas
- Data Science
- Data Pipelines
- Data Visualisation
- Git
- GitHub
Software Developer Intern (Data & AI)
Irys Group
Mar 2026 – May 2026
- Designed and developed web-based, data-driven solutions to improve internal inventory tracking and operational efficiency.
- Translated business requirements into technical specifications.
- Contributed to software development initiatives involving data and AI-focused problem solving.
- Python
- Data Science
- AI
- Web Development
IT Support & Inventory Specialist
Irys Middle East LLC
Sep 2023 – Feb 2025
- Resolved 50+ technical issues monthly and configured systems across 100+ endpoints.
- Delivered training sessions for 20+ staff.
- Improved system performance by 30%.
- IT Support
- Systems Administration
- Inventory Management
Web Developer Intern
RCA Technology
Feb 2023 – Aug 2023
- Developed web features and improved performance by 20%.
- Worked in a team to deliver projects on time.
- Web Development
- Frontend
- JavaScript
Other experience
- Fresh Convenience Sorting Team Member (Part-time)Woolworths Group, MelbourneJun 2025 – Present
Academic Background
Bachelor of Computer Science (Data Science Major)
Deakin University
Jul 2024 – Feb 2027
- Distinction average
Pursuing a Bachelor of Computer Science with a major in Data Science, focusing on software engineering, artificial intelligence, and modern data-driven application development.
- Data Science
- Software Engineering
- Artificial Intelligence
- Machine Learning
Diploma in Information Technology
ANC Education, Colombo, Sri Lanka
Sep 2022 – Sep 2023
- Program score 83 / 100
Completed a diploma covering programming, databases, software development, web design and computer architecture, with High Distinction in 8 of 10 courses.
- Information Technology
- Databases
- Software Development
- Computer Networks
Skills
Languages
- Python
- SQL
- TypeScript
- JavaScript
Data & ML
- Pandas
- Scikit-learn
- LightGBM
- SHAP
- TensorFlow
- PyTorch
Web & Backend
- Next.js
- React
- Flask
- FastAPI
- PostgreSQL
Tools
- Git
- GitHub Actions
- Power BI
- Tableau
- Excel
AI-Assisted Development
- Claude
- Claude Code
- Antigravity
- Codex
- Framer
About Me
I'm a final-year Computer Science student at Deakin, majoring in Data Science, and I like work that ends in something people can actually use. That's why I build end to end: from cleaning messy open data and training forecasting models to shipping the web app on top of it.
Most of my projects start with a real Melbourne question, like how busy the CBD will be tomorrow or how rain changes transport demand, and finish as a live product running on free tiers. At Chameleon I also led a data science team, setting review standards and mentoring teammates while staying the #1 contributor in the repository.
I'm looking for graduate and early-career roles in data science and software engineering, where I can own problems from data to deployment. Outside of that, I'm drawn to startups and entrepreneurship: I enjoy spotting a gap, testing it quickly and building the first version myself. Long term, I want to build products and businesses of my own.
Core Interests
Data Science
Turning messy real-world data into forecasts and insights people can act on.
Software Engineering
Building reliable, maintainable full-stack products, from pipelines to polished UIs.
Tech Entrepreneurship
Spotting gaps, testing ideas fast and building first versions of products and businesses.
Let's Work Together
Let's work together. I'm open to internships, graduate roles, freelance opportunities, and collaborations.