Sunwise

Inspiration

We were inspired by the growing need for sustainable living solutions and the potential of solar energy to reduce electricity costs and environmental impact. With increasing accessibility to satellite data and machine learning, we saw an opportunity to build a tool that empowers individuals to visualize how much they can save—and help the planet—by switching to solar energy.

What it does

Our project, Sunwise, predicts the average solar energy received monthly for any given location using machine learning models. It also analyzes rooftop suitability using image recognition. By entering your rooftop area and monthly electricity bill, the platform visualizes solar potential, estimated savings, carbon offsets, tree equivalent metrics, and roof qualification—all in a clean, user-friendly interface.

How we built it

Machine Learning: Trained a Random Forest model using NASA’s solar data collected from various coordinates in Chicago to estimate solar potential. Computer Vision: Developed a sunroof detection model using transfer learning and PyTorch for image recognition. Frontend & Backend Integration: Built a responsive React.js web app connected via Flask to visualize predictions and provide real-time analytics based on user input.

Challenges we ran into

One of the biggest challenges was collecting and preparing accurate datasets for training our solar potential and rooftop detection models. Another hurdle was integrating the machine learning models into a seamless web application experience using Flask, ensuring the backend remained efficient while communicating with the React frontend.

Accomplishments that we're proud of

We’re proud of building an end-to-end solution that: Combines real-world data and AI in a meaningful way Helps users estimate solar viability for their homes Delivers a functional and responsive UI under the time constraints of a hackathon Integrates machine learning and image recognition into a smooth user experience

What we learned

Throughout this hackathon, we learned how to: Leverage publicly available datasets for impactful insights Train and deploy machine learning models in real-world applications Work with full-stack technologies (Flask, React, PyTorch) Collaborate effectively as a team with diverse technical roles

What's next for Sunwise

We plan to: Expand sunroof detection capabilities to more cities and refine accuracy Add authentication and allow users to save their analysis Integrate real-time satellite imagery and dynamic weather forecasts Partner with solar companies to provide installation estimates or connect users to service providers