πŸŒ† CivicFix

AI-Powered Urban Infrastructure Intelligence Platform


🚨 Problem

Cities struggle to respond efficiently to everyday infrastructure failures β€” potholes, broken streetlights, flooding, unsafe sidewalks, and public space damage.

Current reporting systems:

  • Are slow and manual
  • Lack transparency
  • Don’t prioritize severity effectively
  • Provide little accountability to residents

Consequences:

  • Unsafe streets
  • Unequal service across neighborhoods
  • Loss of public trust

πŸ’‘ Solution

CivicFix is an AI-powered civic reporting and transparency platform that enables residents to report infrastructure issues while giving municipalities real-time intelligence and accountability metrics.

For Residents:

  • Upload a photo of an issue
  • Add a short description
  • Automatically receive AI classification (pothole, streetlight, graffiti, trash) and severity scoring
  • See their report appear instantly on a public dashboard

For Municipal Administrators:

  • View real-time issue heatmaps
  • Track response times
  • Prioritize high-risk hazards
  • Monitor service equity across neighborhoods

πŸ’‘ Inspiration

CivicFix was inspired by everyday problems in our communities β€” potholes left unfixed, broken streetlights making areas unsafe, and reports that disappear without updates.

We wanted a platform that gives people a stronger voice while helping cities work efficiently. By combining AI with real-time data and public dashboards, CivicFix bridges the gap between residents and local government β€” turning complaints into actionable, visible progress.


🎯 Alignment to Challenge Goals

CivicFix contributes to:

  • Public safety through early hazard detection
  • Smarter, data-driven city planning
  • Equitable service across neighborhoods
  • Transparency with public dashboards
  • Community participation

πŸ€– AI Integration

  • Image Classification: Detect issue types (pothole, streetlight, trash, graffiti)
  • Severity Scoring: Prioritize urgent cases
  • Accountability: All AI decisions are logged and stored for auditing

πŸ— Tech Stack

Frontend:

  • React
  • Interactive map visualization (Leaflet / Mapbox)

Backend:

  • Python (FastAPI)
  • AI models: MobileNetV2, Google Gemini 2.5 Flash
  • REST API architecture

Database:

  • PostgreSQL (structured civic data storage)

Infrastructure:

  • Cloud-hosted deployment with scalable architecture

🚧 Major Challenges

  • Selecting AI models suitable for free tiers and lightweight deployment
  • Labeling and scoring issues accurately
  • Ensuring AI correctly identifies various problem types from photos

πŸŽ‰ Accomplishments

  • Implemented AI model fallback for reliability
  • Live map that updates instantly with new reports
  • Public dashboard for transparency
  • Dashboard for municipal staff to prioritize work
  • Demonstrated a real solution for community-driven urban problem solving

πŸ“š Lessons Learned

  • Integrating AI models into real-world features
  • Using a database to store and query geolocation data
  • Designing intuitive interfaces for all users

πŸ“Š Impact

Even as a prototype, CivicFix demonstrates how AI can modernize local governance:

  • Reduce response times
  • Prevent minor hazards from escalating
  • Identify underserved neighborhoods
  • Improve civic engagement

πŸš€ Next Steps

  • AI narrative summaries (e.g., weekly issue reports and budget impact)
  • Data-driven budgeting strategies
  • Budget impact forecasting
  • Equity analytics dashboard

CivicFix is designed to scale from small townships to major metropolitan cities.

Built With

Share this project:

Updates