GEFIFI - Construction Marketplace & Management Platform

GEFIFI is a cutting-edge Marketplace and Project Management platform ecosystem designed to modernize and streamline the construction industry in India. It bridges the gap between Customers, Construction Experts, and Material Suppliers through a unified, transparent, and AI-enhanced digital experience.

🚀 Vision

To empower the Indian construction sector by fostering trust, transparency, and efficiency. GEFIFI simplifies complex project workflows—from finding the right professional to securing quality materials and managing digital contracts.

✨ Key Features

• Multi-Stakeholder Marketplace: Tailored interfaces for Customers (Visionaries), Experts (Builders), and Suppliers (Resource Providers).
• Construction Management: End-to-end tracking of work requests and material procurement.
• AI-Powered Builder Assist: Deeply integrated agentic AI to help users navigate project complexities, manage documents (DWG, PDF, etc.), and facilitate communication.
• Digital Contracts & Agreements: Securely link and manage agreements between parties directly within the platform.
• Real-time Collaboration: Advanced chat system with file sharing, presence tracking, and typing indicators.
• Premium UI/UX: A high-performance, responsive interface featuring modern glassmorphism aesthetics and smooth micro-animations.

🛠️ Technical Stack

Frontend

• Framework: Svelte 5 & SvelteKit (Runes, Snippets)
• Styling: Tailwind CSS 4.0 (Modern Glassmorphism)
• State Management: Svelte's reactive $state and $derived runes.
• Utilities: Wavesurfer.js (Audio/Visual data), Marked (Markdown rendering), Lucide Svelte (Icons).

Backend & Infrastructure

• Runtime: Bun / Node.js with TypeScript
• Backend as a Service: Firebase (Auth, Firestore, Functions, Storage, Hosting)
• Deployment: Google Cloud Platform (GCP) with Cloud Build.
• Containerization: Docker for consistent environments across development and production.

AI Engine

• Framework: Developed using the ADK (Agent Development Kit) in Python for robust agent orchestration.
• Agents: Custom agentic workflows integrated with Google's Gemini models.
• Capabilities: Document parsing (DWG/DXF/PDF), automated session management, and project coordination assistance.

🛠️ Getting Started

Prerequisites

• Bun or Node.js
• Firebase CLI

Installation

1. Clone the repository:

   git clone https://github.com/your-username/gefifi-2.git
   cd gefifi-2
   

2. Install dependencies:

   bun install
   

Running Locally

To run the full stack locally for development, open four separate terminal tabs and run the following commands in order:

1. Tab 1 (Firebase Emulators):

   ./scripts/firebase-emulator-start.sh
   

2. Tab 2 (Backend Services):

   ./scripts/run-backend.sh
   

3. Tab 3 (AI Agents):

   ./scripts/run-agents.sh
   

4. Tab 4 (Frontend App):

   bun run dev
   

🔒 Security

The project implements rigorous security measures through Firestore Security Rules, ensuring:

• Data privacy for personal profiles and contracts.
• Participant-based access control for chats and project management nodes.
• Role-based permissions for Customers, Experts, and Suppliers.

📝 License

This project is private and proprietary. All rights reserved.