Redcross CrisisSync

Submit Report
Overview -dashboard
Fusion Queue
Report Status
Cluster details

Inspiration

During disasters and humanitarian emergencies, the International Federation of Red Cross and Red Crescent Societies (IFRC) and National Societies have to deal with large amounts of information about the needs of affected communities. Field reports can be unstructured, incomplete, and difficult to process quickly.

We were inspired by the idea of using AI to help the humanitarian response ecosystem turn these reports into clear, structured information about essential needs. Our main goal with RedCross CrisisSync is to support the IFRC's humanitarian mission by making the process of understanding and organizing needs faster and more efficient.

What it does

RedCross CrisisSync is an AI-powered humanitarian needs assessment and coordination platform.

Users can submit reports describing the conditions and requirements of affected communities. The system uses AI to analyze these reports and identify essential needs such as:

Food
Water
Shelter
Medical assistance
Sanitation
Clothing
Other essential relief requirements

The extracted information is presented in a structured format, helping humanitarian teams understand reported needs more efficiently.

Our core workflow is:

$$ \text{Humanitarian Report} \rightarrow \text{AI Analysis} \rightarrow \text{Need Extraction} \rightarrow \text{Structured Information} $$

The broader vision is to develop this into a solution that can support IFRC and Red Cross/Red Crescent National Societies in humanitarian information management and response coordination.

How we built it

We built RedCross CrisisSync as a full-stack web application.

Frontend

React
TypeScript
Vite
Tailwind CSS

Backend

Python
FastAPI
Pydantic
Uvicorn

AI

Google Gemini API

The React frontend provides the user interface, while the FastAPI backend manages application logic, REST APIs, data processing, and communication with the Gemini API.

We designed the system so that a humanitarian report can move through the complete pipeline—from submission to AI analysis and structured presentation—within a single platform.

Challenges we ran into

One of our biggest challenges was dealing with the ambiguity and variability of humanitarian reports. Real-world reports may contain incomplete information, different descriptions of the same need, or information that requires human interpretation.

Another challenge was ensuring that AI-generated results are useful without treating them as unquestionable facts. This made us think carefully about the role of AI in humanitarian applications: AI should assist humanitarian personnel, not replace their judgment.

We also faced technical challenges while integrating a React frontend, FastAPI backend, REST APIs, and the Gemini AI service into one working system.

Accomplishments that we're proud of

We are proud of turning an idea focused on humanitarian information overload into a working AI-powered full-stack platform.

Our key accomplishments include:

Building a functional web-based humanitarian needs assessment system.
Integrating Google Gemini into a real-world humanitarian workflow.
Connecting a React/TypeScript frontend with a Python/FastAPI backend.
Creating a structured process for extracting essential needs from reports.
Designing the project with the IFRC and Red Cross/Red Crescent humanitarian ecosystem as the long-term target.
Demonstrating how generative AI can be applied beyond conversational interfaces to support humanitarian information management.

Most importantly, we are proud that the project focuses on a meaningful problem where better information processing can potentially support people affected by disasters.

What we learned

RedCross CrisisSync taught us that building an AI application is not simply about connecting an API to a user interface.

We learned how to:

Integrate generative AI into a practical workflow.
Build and connect a modern React frontend with a FastAPI backend.
Design REST APIs for communication between application components.
Process unstructured information into structured data.
Think about AI reliability, ambiguity, and human verification.
Design technology around the needs of a real-world humanitarian ecosystem.
Consider how a prototype can evolve toward a larger-scale platform suitable for organizations such as IFRC.

Most importantly, we learned that responsible AI in humanitarian settings must keep humans in the loop.

What's next for RedCross CrisisSync

Our next goal is to move RedCross CrisisSync from a prototype toward a more robust humanitarian information platform aligned with the needs of the IFRC and Red Cross/Red Crescent National Societies.

Future development could include:

More robust humanitarian needs classification.
Multilingual and voice-based report submission.
Improved handling of incomplete and conflicting reports.
Stronger validation and human-review mechanisms for AI-generated results.
Secure authentication and role-based access.
Persistent, scalable data storage.
Integration with existing humanitarian information systems and workflows.
Analytics to help identify changing patterns in reported needs.
Evaluation with realistic humanitarian datasets and, where possible, feedback from humanitarian practitioners.

Our long-term vision is for RedCross CrisisSync to become a reliable AI-assisted layer for humanitarian needs information—helping IFRC and its network process critical information faster while keeping humanitarian expertise and decision-making at the center.

Built With

ai/ml
api-integration
bycrypt
fastapi
gemini-api
javascript
npm
pydantic
pytest
python
react
restapi
tailwindcss
typescript
uvicorn
vercel
vite

Submitted to

Beginner's Paradise - FirstCommit

October 18 at 5:00pm EDT to vote

Vote for this project!
Hyperbloom September - Beginer Freindly
InfinityX Global Hackathon 2K26
LovHack Season 3
CodeSprint by Elite Coders
InternWell Hack2.0

Created by

Contributed to the UX/UI design of the Project Red Cross Crisis Sync and supported the team with backend development and minor implementation tasks, helping improve the overall functionality and user experience of the project.

Prarthana A Nair
I contributed to RedCross CrisisSync primarily through backend development, API integration, and technical research. I worked on understanding the system requirements and translating the identified humanitarian challenges into practical backend functionality. I contributed to developing and testing APIs, connecting backend services with the frontend, and troubleshooting issues that arose during development.

I also worked on debugging and improving the reliability of the application, including investigating API and report-related issues and ensuring that the backend could properly support the features implemented in the frontend.

Through this project, I gained practical experience in backend development, API integration, debugging, and collaborative software development. More importantly, I learned how effective technology solutions for real-world humanitarian challenges require both a clear understanding of the problem and a reliable technical foundation.

Rafa Fathima M K
I contributed to RedCross CrisisSync primarily through database design and Data/AI development. I designed and implemented the PostgreSQL schema for field reports, needs, priority, locations, evidence, and clusters, along with Row-Level Security (RLS) policies to manage data access at the database level.

I also developed the AI-facing data layer, including an ExtractionService using Google Gemini to extract structured needs and summaries from raw field reports, and a retrieval-based AI assistance service that retrieves relevant information from stored reports to generate context-based responses for responders.

I was also involved in integration and debugging, verifying the pipeline from AI extraction through PostgreSQL repositories and retrieval, and merging teammates’ pull requests to integrate their work.

Through this project, I gained practical experience in PostgreSQL schema design, RLS, backend development, LLM integration, retrieval-based AI systems, Git collaboration, and deployment workflows. The project also helped me understand the importance of structuring and retrieving data effectively when building AI-assisted systems for humanitarian applications.

Sahadiya fathima M
I mainly worked on the frontend development and research for RedCross CrisisSync. I contributed to identifying and understanding the problem we wanted to solve, researching the humanitarian needs and challenges faced during disaster response, and exploring possible solutions that could realistically address them. I also worked on designing and developing the frontend, focusing on creating a clear, user-friendly interface and connecting the different parts of the application into a smooth user experience. Through this project, I learned how important it is to combine proper problem research with practical technology when building a solution for a real-world humanitarian challenge.

AISHWARYA NAIR