ArcLight AI 2.0 — Urban Intelligence OS

Real-time city sustainability intelligence platform powered by AI, IoT sensors, and open data APIs.

🌐 Live: https://arclight-ai.vercel.app

🌆 Overview

ArcLight AI 2.0 is a production-grade Urban Operating System that unifies city-wide data streams into a single, intelligent dashboard. It brings together Waste Management, Energy Optimization, Air Quality Monitoring, and Citizen Engagement — all in real-time.

Built with a premium white glassmorphism UI, it's designed to look and feel like the future of smart city governance.

✨ Features

Module	Description
🗑️ Smart Waste	Computer vision fill-level detection, AI-optimized collection routing, 6-hour predictions
⚡ Smart Energy	Occupancy-based HVAC control, solar optimization, anomaly detection
🌐 Digital Twin	3D city simulation, festival scenario planning, real-time IoT sync
🤖 AI Copilot	RAG-grounded LLM with real-time sensor context, multi-provider fallback
👥 Citizen Layer	Gamified reporting, YOLOv11 CV verification, community leaderboard
💨 AirSense	Live AQI from WAQI API, PM2.5/PM10/NO₂ trends, 7-day weather forecast
📊 Command Center	Unified KPI dashboard, city-wide sustainability score, live activity feed

🛠️ Tech Stack

Frontend: Next.js 15, TypeScript, Recharts, Lucide Icons
Styling: Custom Glassmorphism CSS (no Tailwind required for glass effects)
AI: Gemini 2.5 Flash → Groq (Llama 3.1) → OpenAI GPT-4o Mini (fallback chain)
Real-Time Data: Open-Meteo API (weather & solar), WAQI API (air quality)
Simulation: Physics-based models coupled to live weather data

🚀 Getting Started

1. Clone the repository

git clone https://github.com/Rishabhv16/arclight-ai.git
cd arclight-ai

2. Install dependencies

npm install

3. Set up environment variables

Copy the example file and fill in your API keys:

cp .env.example .env.local

4. Run the development server

npm run dev

Open http://localhost:3000 to run locally, or visit the live demo at https://arclight-ai.vercel.app.

🔑 Environment Variables

See .env.example for all required variables.

Variable	Description	Source
`GEMINI_API_KEYS`	Comma-separated Gemini API keys	Google AI Studio
`GROQ_API_KEYS`	Comma-separated Groq API keys	Groq Console
`OPENAI_API_KEY`	OpenAI API key (fallback)	OpenAI Platform
`WAQI_TOKEN`	World Air Quality Index token	aqicn.org

Command Center Dashboard

Real-time KPIs, energy graphs, waste distribution, and live activity feed.

AirSense Module

Live AQI data from real WAQI monitoring stations in Hyderabad with 24-hour trends.

📁 Project Structure

arclight-ai/
├── app/
│   ├── page.tsx              # Landing page with Unsplash hero
│   ├── dashboard/page.tsx    # Command Center
│   ├── airsense/page.tsx     # AirSense module (WAQI + Open-Meteo)
│   ├── waste/page.tsx        # Smart Waste module
│   ├── energy/page.tsx       # Smart Energy module
│   ├── copilot/page.tsx      # AI Copilot
│   ├── citizens/page.tsx     # Citizen Layer
│   ├── twin/page.tsx         # Digital Twin
│   ├── api/chat/route.ts     # AI provider fallback chain
│   └── globals.css           # Premium glassmorphism design system
├── components/
│   ├── layout/               # AppShell, Sidebar, TopBar
│   ├── charts/               # Recharts wrappers
│   └── ui/                   # MetricCard, AlertBanner
└── lib/
    ├── api.ts                # Open-Meteo & WAQI data fetching
    └── simulators/index.ts   # Weather-coupled physics simulations

🌍 Real Data Sources

Weather: Open-Meteo — Free, no API key required for basic weather
Air Quality: WAQI API — Real monitoring stations (Bollaram, Pashamylaram, ECIL, IDA Jeedimetla, Hyderabad)
Images: Unsplash — Free high-quality photography