Projects
From ideas to real-world solutions. Here's a collection of projects where I've explored data, built models, and turned complex problems into practical solutions. Each project has taught me something new, challenged the way I think, and given me the opportunity to put my skills into practice.
Renery — Smart Energy Planner
Currently in Development
Master's Capstone Project | Monash University
Renery is a web application developed as part of my Master's capstone project at Monash University. It helps Australian households find greener and more cost-effective times to use electricity by combining renewable-energy forecasts, electricity tariffs, and personal schedules.
The application supports activities such as washing, dishwashing, clothes drying, and EV charging.
Problem
Renewable-energy availability and electricity prices change throughout the day, but households often lack a simple way to compare this information and decide when to use electricity.
Tools
Python, Pandas, Scikit-learn, PostgreSQL, React, Tailwind CSS, Node.js, Express.js, AEMO Energy Data, AER Electricity Plan Data.
Outcome
Built a data pipeline to collect, clean, and manage Australian energy data. Worked with the team to develop renewable-energy forecasts, smart appliance scheduling, and environmental impact tracking.
Spotify Data Analytics
Built a comprehensive data analysis pipeline covering data collection, cleaning, database design, statistical analysis, and machine learning. The project explored audio features and user behaviour patterns across multiple countries and time periods, enabling insight generation and visualization.Problem
Understand patterns in music data and identify factors influencing song popularity and user listening behaviour.
Tools
Python, SQL, Pandas, Machine Learning (Regression, Clustering, Classification), Power BI.
Outcome
Developed an end-to-end analytics pipeline and interactive dashboards to generate insights and support data-driven recommendations.
Chronic Disease Prediction & Progression Analysis
Analyzed anonymized patient health data to study chronic disease risk and progression using both exploratory analysis and predictive modeling. The project examined how variables such as smoking, diet quality, gender, biomarkers, and lifestyle factors relate to disease outcomes, while also building interpretable classification models for chronic disease prediction.Problem
Analyze how demographic and lifestyle factors influence chronic disease risk and identify meaningful patterns across patient populations.
Tools
R, tidyverse, ggplot2, caret, recipes, logistic regression, multinomial logistic regression.
Outcome
Combined exploratory healthcare analytics with predictive modeling to identify disease-related patterns and support data-driven interpretation.
Amazon Laptop Data Analysis
Collected real-world product data through web scraping, performed data cleaning and feature engineering, and developed machine learning models to predict laptop prices and analyze feature importance across different product categories.Problem
Analyze laptop pricing and specifications to understand market trends and key pricing factors.
Tools
Python, Selenium, BeautifulSoup, Pandas, Linear Regression, Random Forest.
Outcome
Built predictive models to estimate laptop prices and evaluate how technical specifications affect pricing.
