CivicPulse-AI

Inspiration

Cities are filled with everyday civic issues — potholes, garbage overflow, broken streetlights, and sewage leaks — that often go unnoticed until they become serious problems. We were inspired by the need for a faster, smarter, and more transparent way for citizens and local authorities to report and resolve these issues in real time.

What it does

CivicPulse-AI is an AI-powered civic issue triage platform that allows citizens to submit reports with photos, GPS location, and issue details. The system uses computer vision to detect and classify the problem, evaluates severity and priority, and helps authorities manage incidents efficiently through a live dashboard.

The platform combines:

AI image analysis for civic defects Geo-aware duplicate detection Real-time ticket updates using Socket.io A citizen reporting interface An authority command center for monitoring and dispatching work

How we built it

We built CivicPulse-AI as a modular monorepo with three main services:

AI Vision Service using Python, FastAPI, OpenCV, PyTorch, and Ultralytics YOLOv8 for defect detection and severity scoring Backend Service using Node.js, Express, MongoDB, and Socket.io for ticket management and live event streaming Frontend App using React, Vite, Tailwind CSS, and Leaflet for the citizen reporting experience and command center dashboard The workflow is designed to be efficient:

A citizen uploads an image and location The AI model analyzes the defect and assigns a priority score The backend checks for nearby duplicate incidents A ticket is created or merged Authorities receive real-time updates and can assign or resolve the issue.

Challenges we ran into

One of the biggest challenges was integrating the AI analysis pipeline with the backend and frontend in a way that remained reliable and responsive. We also had to handle:

dependency mismatches across Python and Node packages local environment setup without external services building a fallback workflow when MongoDB or AI services are unavailable designing a real-time event system that updates the dashboard without requiring page refreshes

Accomplishments that we're proud of

We’re proud to have created a complete working prototype of a civic issue intelligence platform that combines:

AI-powered defect detection real-time city operations monitoring duplicate report prevention interactive geospatial dashboards a clean and functional user experience for both citizens and authorities

What we learned

We learned how important it is to design systems with modularity and resilience in mind. Combining AI, backend APIs, and live frontend updates requires careful data flow planning, dependency management, and graceful fallback handling. We also gained valuable experience in building full-stack systems that connect real-world user input with operational decision-making.

What's next for CivicPulse-AI

We plan to expand CivicPulse-AI with:

better model accuracy for more civic defect categories live worker assignment and route optimization analytics dashboards for city departments multilingual reporting support optional integration with real municipal data sources and public APIs

Built With

• express.js
• fastapi
• leaflet.js
• lucide
• mongodb
• node.js
• opencv
• python
• pytorch
• react
• socket.io
• tailwind-css
• ultralytics-yolov8
• vite