LOG // PROJECTS
Projects
Real problems. Real solutions. Click any card to see the full story.
AI-Powered Exoplanet Discovery
NASA Space Apps
Event: NASA Space Apps 2025
Difficulty: Advanced
Duration: 36 Hours
The Challenge
For the 2025 NASA Space Apps Challenge, we chose the advanced "A World Away: Hunting for Exoplanets with AI" challenge. The problem was clear: while missions like Kepler, K2, and TESS have enabled the discovery of thousands of exoplanets, most were identified through time-consuming manual analysis. Our task was to create a machine learning model, trained on NASA's open datasets, that could automatically and accurately identify new exoplanets from raw astronomical data.
Our Solution
My team and I decided to go a step further. Instead of only developing a model, we built a full-stack interactive web platform to empower both researchers and citizen scientists. The platform allows users to upload their own astronomical data for evaluation by our models, train new models, and visualize the relationship between a celestial body and its host star.
My Role & Contributions
As a core developer, I architected the full-stack application, leading the frontend development with Next.js and building the Python/Flask backend API to serve our machine learning models. I was responsible for the seamless integration between the complex data science backend and the intuitive user interface.
Tech Stack
Python Flask Scikit-learn Next.js Pandas
