Data Analysis / Hackathon Project
Weatherly
NASA hackathon project comparing environmental comfort between locations with AI-assisted insights.
Scroll to explore the build ↓
Case study
Weatherly
Built around performance, scalability, and pragmatic engineering choices.
Overview
A weather analysis tool developed for a NASA hackathon that evaluates environmental comfort between two locations.
Tech stack
Gallery
The problem
Travelers and event planners lacked an intuitive way to compare comfort between two destinations beyond raw weather metrics.
The solution
Weatherly consumed NASA POWER data, computed derived comfort metrics, and summarized them in a way that humans can quickly understand. An AI assistant added narrative explanations on top of the numbers.
Technical decisions
Python, Pandas, and NumPy handled the statistical side; Jupyter helped iterate on metrics. The architecture stayed lean to focus on analysis and visualization rather than infrastructure.
What I learned
I learned how to translate dense scientific datasets into human-centered stories, and how small, focused models can provide real value in niche decision-making flows.