Inspiration

When someone goes missing, information can come from family members, friends, witnesses, and volunteers through different channels. Organizing these reports quickly can be challenging.

We were inspired by the idea of using AI to organize scattered reports into structured information and support faster rescue coordination. This led us to build RescueLink, an AI-powered Missing Person Rescue Coordinator.

What We Built

RescueLink allows users to create missing-person cases and collect sighting reports from the community.

The system uses AI to:

Extract information from sighting reports
Structure details such as location, time, clothing, and clues
Prioritize reports for human review
Organize sightings into a timeline
Generate an AI-assisted rescue briefing
Support English, Tamil, and Hindi reports

The platform also includes an interactive map for viewing the last known location and reported sightings, along with volunteer coordination.

How We Built It

We built RescueLink using:

React + Vite for the frontend
Python + Flask for the backend
SQLite for the database
Google Gemini API for AI processing
Leaflet + OpenStreetMap for map visualization

AI is used to organize and prioritize information for human review rather than confirm a person's identity.

We also keep the Gemini API key on the backend using environment variables.

What We Learned

Building RescueLink helped us learn how to:

Build a full-stack AI application
Integrate the Gemini API with Flask
Work with structured AI responses
Handle incomplete information and API failures
Visualize location-based data
Design AI systems with human oversight
Consider privacy and responsible AI

Challenges We Faced

One challenge was ensuring that the AI does not invent information when analyzing sighting reports. We designed it to return unknown or null values when information is unavailable.

Another challenge was ensuring that AI does not treat a sighting as a confirmed identification. AI-generated relevance is therefore used only to assist human review.

We also considered multilingual reports, API failures, privacy, and usability during stressful situations.

Future Vision

We envision RescueLink as a secure rescue coordination platform that could work with verified authorities, NGOs, emergency organizations, and volunteer networks.

Future improvements could include real-time notifications, stronger privacy controls, offline support, advanced geospatial analysis, and integration with authorized rescue organizations.