Collector

• 

  Identifying buildings with untapped rainwater potential for smarter and more efficient water use.

Inspiration

We were inspired by how easy it has become to build something useful and make a real difference. We wanted to create a tool that could find buildings where rainwater collection would save money and water. The idea of using data and automation to help with sustainability pushed us to build Collector.

What it does

Collector finds buildings that are good candidates for rainwater collection systems. It looks at roof size and other building features using computer vision. Then it estimates how much water can be collected and how much money can be saved. This helps identify strong prospects for water reuse solutions.

How we built it

We built Collector using Python. We used computer vision to analyze building and roof data. We connected datasets and processed them to detect large roofs and other useful features. We also added Auth0 for user authorization so users can securely access the system.

Challenges we ran into

One of the biggest challenges was time. We had to learn new tools quickly and make decisions fast. Computer vision was also difficult because we had to figure out how to work with new datasets. It was overwhelming to choose between many AI tools and decide which ones to use. We also had to adjust our approach as we learned more.

Accomplishments that we're proud of

We successfully implemented computer vision for detecting buildings. We connected datasets and created a working prospect detection pipeline. We also added authentication and built a full prototype in a short time.

What we learned

We learned how to use computer vision with real datasets. We learned how to combine multiple data sources into one system. We also learned how to use Auth0 for authentication and how to quickly evaluate AI tools for solving problems.

What's next for Collector

Next we want to improve detection accuracy and train better models. We plan to add more datasets and improve scoring for water savings and cost savings. We also want to scale the system to cover more locations and provide better prospect insights.

Built With

• javascript
• nextjs
• postgis
• postgresql
• python
• react
• typescript