GoSaveMe - ChatBot and Submission dashboard done with Inertia (Vite & Laravel)

Developed for the DeaconCode Hackathon at Wake Forest University – Best Project Winner

Our project was built during the DeaconCode Hackathon at Wake Forest University, where we were asked to tackle real-world challenges in the health field. We developed a solution that streamlines donation requests for health crises by integrating a dynamic chatbot with a submission dashboard. Our innovative approach earned us first place and recognition for the Best Project.

Project Overview

This project is envisioned to enable non-profits to manage donation requests for health crises securely and efficiently. Unlike traditional donation platforms like GoFundMe, our solution prioritizes privacy and fraud verification by collecting sensitive information through an interactive chatbot without publicizing the applicant’s health crisis.

Key Challenges Addressed

• Efficient Donation Request Submission:
  The platform uses a chatbot powered by the Gemini API to ask dynamic, case-specific questions. Gathering detailed information required to verify the case.

• Submission Dashboard:
  A dashboard lists all submissions and provides expandable views for administrators to review and verify each request.

Technologies Used

• Frontend: Inertia.js, React, Vite, shadcn UI components
• Backend: Laravel
• ChatBot Integration: Gemini API
• Styling & Icons: Tailwind

Getting Started

Prerequisites

• Node.js
• PHP
• Composer
• Gemini API

Installation

1. Clone the repository:

2. Install backend dependencies:

   composer install
   php artisan key:generate

3. Install frontend dependencies:

   npm install
   npm run dev

4. Configure your environment:

   • Copy ".env.example" to ".env" and update your database configuration.
   • Set your Gemini API key in your environment file:

     VITE_GEMINI_API_KEY=your_gemini_api_key_here
     

5. Run database migrations:

   php artisan migrate

6. Run the Project

php artisan serve
npm run dev

Future Directions

We plan to further enhance the verification process using machine learning techniques to automatically detect and classify false submissions. Contributions and suggestions to evolve this platform are welcome.

Team:

• Alejandro Gonzalez Rubio
• Saroj Bhatta
• Risal Shahriar