CivicLens

Inspiration

Cities are full of problems that go unnoticed—potholes, flooding, broken streetlights—but reporting them is often slow, confusing, or ignored. At the same time citizens what to be heard and have real influence on their neighborhood. We wanted to create a platform that empowers people to report infrastructure issues quickly and without a lot of effort, while giving city administrators the tools to prioritize and respond in real-time. CivicLens was born from the idea of combining community action, AI, and real-time technology to make cities safer, smarter, and more connected.

What it does

CivicLens is a civic reporting platform where users can use a mobile app to snap a photo, add context, and report local issues. Our AI, automatically identifies the type of issue, finds the correct department to submit the issue and rank the severity of the issue. Administrators use a real-time dashboard built with Firebase Firestore, which shows critical cases, and live reactions, allowing them to quickly respond and keep the community informed. Citizens receive real-time notifications as their reports move from “pending” to “resolved,” strengthening trust and collaboration.

How we built it

CivicLens is a full-stack mobile application with a monorepo structure:

Frontend: React (Vite) + Tailwind CSS, designed mobile-first, featuring a citizen feed and admin dashboard. Users can upload photos, submit reports, and react to community posts.

Backend: Node.js + Express, connected to Firebase Firestore for real-time database management. We built APIs for adding reports, updating status, handling reactions, and streaming notifications.

AI Pipeline: Integrated the Claude Vision API to classify issues, summarize reports, and calculate severity scores, combining AI insights with community input for smarter prioritization.

Challenges we ran into

Designing the LLM vision pipeline to accurately classify diverse infrastructure issues from images.

Synchronizing real-time updates between Firebase, the citizen feed, and the admin dashboard.

Balancing severity scoring between AI-assessed danger, reaction counts, and time-based urgency.

Creating a smooth mobile-first UI that handles image uploads, geolocation, and notifications seamlessly.

Accomplishments that we're proud of

We successfully built a demo-ready CivicLens app where citizens can report issues, see AI-generated summaries, and get notifications as their reports are addressed. Administrators can track cases by severity, see live reactions, and visualize problems on a map. CivicLens bridges technology and community, empowering citizens and governments to collaborate in real-time for safer, healthier cities.

What we learned

We learned how to build a full-stack, real-time app connecting a React frontend, Node.js backend, and Firebase database. We applied AI to classify issues and prioritize them with severity scoring. We also gained experience designing for both citizens and admins, handling real-time updates, and rapidly prototyping a scalable, user-centered civic solution.

What's next for CivicLens

Expanding to other New Jersey cities beyond Newark, adding a live map view showing all active reports by severity, integrating weather data to dynamically boost severity scores for weather-related issues, building a native iOS/Android app via React Native, and partnering with actual Newark municipal departments for a real pilot program.