Projects

Ensemble Rainfall Prediction Model

UBC (Mar 2025)
Designed an ensemble machine learning model for daily rainfall prediction using Australian climate datasets.
Deployed a Flask REST API on AWS EC2, using S3 and EMR Spark for scalable data handling.

---

Ads Analytics Dashboard

UBC (Feb 2025)
Developed an interactive web dashboard (Dash + Flask) to analyze Google Play Store app data for advertisers.
Enabled filtering by installs, categories, and ratings, with visual comparisons using Altair.
Integrated caching, production server deployment (Gunicorn), and live interactivity.
GitHub Repository

---

Datpro – Data Profiling Package

UBC (Jan 2025)
Built a modular Python package for automated data cleaning, anomaly detection, and EDA visualization.
Supported plots (boxplots, heatmaps, histograms) using Altair; included anomaly detection via IQR and percentile-based methods.
Integrated CI/CD using GitHub Actions, pytest for testing, and ReadTheDocs for deploying the package documentation.
GitHub Repository

---

Age Group Predictor

UBC (Dec 2024)
Created a containerized ML pipeline (Docker + Makefile) to classify age groups from NHANES survey data.
Designed modular architecture aligned with MLOps principles, enabling reproducible training and deployment.
GitHub Repository