Why not make it as easy as taking a photo to get these issues noticed and fixed? Our project turns one simple photo into action. When someone uploads a picture, the system detects what kind of problem it is, like a pothole, graffiti, or fallen tree, figures out where it is using location data, and automatically files a fix request. Then it goes a step further by creating a social media post that helps bring public attention to the issue.
We built the system using FastAPI for the backend that processes images, a Java-based service to handle location and tracking details, and a Supabase database to store everything. For detecting issues, we trained a computer vision model that recognizes different types of city damage. We connected everything through Supabase Edge Functions so that once the issue is identified, a short post gets created and published directly to Twitter using the Twitter API. It’s a simple flow that connects AI, data, and social media in a meaningful way — one photo turns into awareness and, hopefully, a solution.
Setting it up wasn’t easy. We ran into problems making all the APIs talk to each other, and getting the serverless functions to run smoothly took a lot of debugging. Twitter’s API had its own limits, and we had to find workarounds for posting. But by the end, we had a system that could take a real photo of a city issue, identify it correctly, generate a fix request, and post about it online in just seconds.
We’re proud that it actually works and that it can make a difference. The idea is simple but powerful — giving people the ability to improve their city with just one photo. It’s not just about automation; it’s about awareness and civic action, using technology to make communities better, one post at a time.
Built With
- brightdata
- fastapi
- groq
- javascript
- nextjs
- playwright
- python
- react
- supabase
Log in or sign up for Devpost to join the conversation.