VenueSync AI — Predictive Digital Twin & Crowd Load-Balancer

VenueSync AI is an enterprise-grade spatial digital twin designed to actively monitor, manage, and load-balance large-scale venue crowds in real-time. Powered by Next.js 15, Mapbox GL, and the Gemini 2.0 Flash cognitive engine, the system utilizes artificial intelligence to calculate optimal routes, bypass dynamic bottlenecks, simulate IoT environmental feeds, and instantly enact systemic evacuation protocols.

---

⚡ Key Features

🧠 Gemini AI Routing Engine

• Conversational Interface: Users engage with the AI natively in natural language to request routing to exits, merch stands, food courts, or general safety zones.
• Closed-Loop Load Balancing: When the AI computes and assigns a route, the system updates its internal state to reflect the impending influx (pending_arrivals). The Gemini engine intelligently avoids directing new traffic into newly created bottlenecks.
• 3D Constraints (Z-Axis penalties): Nodes contain exact elevation_level data. The AI intrinsically factors in physical climbing penalties (stairs/escalators), steering users toward visually longer—but physically faster—flat routes when appropriate.

🗺️ Real-Time Spatial Digital Twin

• Dynamic Pathfinding Cinematics: Visual Mapbox integration smoothly flies the camera to targeted nodes while animating a dual-layer cyan path tracking from the user's origin to the destination.
• Multi-Event Context Switching: The platform scales globally. An interactive Command Module drop-down instantly sweeps the viewport across the globe (e.g., Chandigarh Sports Complex, Sector 17 Plaza, Exhibition Ground) to monitor disparate geographic events simultaneously.
• Micro-Animations & IoT Hardware: Pulsing radar loops denote zone density thresholds. Absolute top-level layout tracking monitors mocked live feeds of LiDAR, Thermal, and Ticketing Edge sensors.

🚨 Autonomous Panic Propagation ("Evacuation Mode")

An adjustable Venue Stress Level actively monitors event dynamics:

• Upon eclipsing the critical 2.0 threshold, the system tint-shifts immediately to a deep emergency red.
• Map markers execute aggressive pulse frequencies.
• The master Gemini heuristic is forcefully overridden: regardless of the user's intent (e.g., "I want to buy a t-shirt"), the AI rigidly denies the request and enforces an immediate, priority route to the nearest structural exit.

---

💻 Tech Stack

• Framework: Next.js 15 (App Router)
• Language: TypeScript
• Styling: Tailwind CSS v4
• Animations: Framer Motion
• Mapping/GIS: Mapbox GL JS v3 & react-map-gl
• LLM Cognitive Core: Google Gemini 2.0 Flash

---

🛠️ Setup & Installation

1. Clone the repository

git clone https://github.com/krish57-bit/venuesync-ai.git
cd venuesync-ai

2. Install dependencies

npm install

3. Configure Environment Variables Create a .env.local file at the root of the directory and include:

NEXT_PUBLIC_MAPBOX_TOKEN=your_mapbox_public_token
GEMINI_API_KEY=your_gemini_api_key

4. Start the Development Server

npm run dev

---

🎨 Enterprise UI/UX Architecture

Vigorously modeled after top-tier design systems like Linear and Vercel, the layout is deeply rooted in modern web application aesthetics:

• Glassmorphism & Hud Design: Elegant sheer bg-white/[0.02] backgrounds heavily blurred with backdrop-blur-2xl.
• Kinetic Physics: Chat bubbles effortlessly pop into view using raw physics-based spring animations (stiffness: 400, damping: 25).
• Responsive Hybrid Layout: On desktop, the user engages in a 60/40 cinematic split-screen. On smartphones, the layout structurally refactors via Flexbox, shifting the Chat into a full-screen overlay to prevent map crushing on small viewports.

---

Built for Hackathon Demonstration.