🧠 Futuristic RAG Applications & Smart City Digital Twin

An interactive, premium web application built to visualize cutting-edge Retrieval-Augmented Generation (RAG) use cases and the 5-layer architecture of a Smart City Digital Twin.

✨ Features

• Interactive Nodes: Beautiful, glassmorphic UI built with reactflow to explore complex system architectures.
• Dual Views: Seamlessly toggle between:
  • 🌐 RAG Overview: A radial mind map of 7 futuristic AI personas (Doctor Assistant, Legal Advisor, Smart City Twin, Research Copilot, Personal Memory OS, Finance Strategist, Space Mission Explorer).
  • 🏙️ Smart City Deep Dive: A top-down pipeline mapping the 5-layer ingestion, routing, and response architecture for a Smart City Digital Twin.
• Fluid Animations: Modal popups powered by framer-motion for a smooth, app-like feel.
• Quick Copy: View and instantly copy the master system prompts for each agent or layer directly to your clipboard.

🛠️ Tech Stack

• Framework: React.js + Vite
• Visualization: React Flow (reactflow)
• Animations: Framer Motion
• Icons: Lucide React
• Styling: Vanilla CSS (Custom dark theme, glassmorphism)

🚀 Getting Started

Prerequisites

Make sure you have Node.js installed on your machine.

Installation

1. Clone the repository:

   git clone git@github.com:OracleBrain/neura-rag.git

2. Navigate into the project directory:

   cd neura-rag

3. Install the dependencies:

   npm install

Running Locally

Start the Vite development server:

npm run dev

Navigate to http://localhost:5173 in your browser to view the application.

🏗️ Architecture Visualization

1. RAG Personas Overview

A central "RAG Engine" node connecting outward to specialized AI assistants. Each node contains a highly detailed prompt ready to be deployed into a vector-database-backed LLM pipeline.

2. Smart City Pipeline (5-Layer)

• Master System Prompt: The core orchestrator.
• Layer 1 (Data Ingestion): Extracting structured facts from sensors/GIS.
• Layer 2 (Query Routing): Classifying queries (traffic, energy, emergency).
• Layer 3 (Core RAG): Generating answers grounded in live sensor data and retrieved chunks.
• Layer 4 (Emergency Alert): Triggered automatically on critical thresholds.
• Layer 5 (Urban Planning): Evaluating long-term decision support and policies.

📝 License

This project is open-source and available under the MIT License.