V2.0 Core Intelligence Online

ArenaOS AI
for Smart Venues.

ArenaOS AI combines live operational telemetry, spatial visualization, and AI‑assisted incident response into a unified command experience.

Built with
Next.jsReact Three FiberThree.jsFastAPIOpenCVNVIDIA NIM
Scroll
The Problem

Traditional operations are fragmented

Venues today run on dozens of disconnected screens. When an incident happens, operators piece together the situation manually — losing critical seconds.

Legacy Systems
  • Siloed CCTV grids with no spatial context
  • Radio communication causing response lag
  • Fragmented BMS, access control, fire dashboards
  • Reactive decision-making under pressure
The ArenaOS Paradigm
  • One unified 3D command center for all systems
  • AI copilot with numbers-driven incident analysis
  • Live interactive digital twin synced to reality
  • Proactive, predictive situational awareness
Platform

Everything in one system

Every subsystem is intrinsically linked to the core intelligence engine.

Live Digital Twin

A physically accurate 3D stadium mapped to real-time sensor telemetry. Every occupancy change, every zone alert, rendered live.

AI Copilot

Analyzes structured incident data and generates operational recommendations. Supports optional LLM integration for advanced contextual reasoning.

Computer Vision Pipeline

Supports webcam‑based occupancy estimation today and is architected for future RTSP/IP‑camera integration.

Predictive Analytics Prototype

Demonstrates how AI‑driven crowd forecasting can be integrated into the operational workflow.

Multi‑Agent UI

Visualizes collaborative AI agents responsible for monitoring, analysis, planning, and response. The architecture supports future autonomous orchestration.

Drone Dispatch

Visualizes coordinated drone deployment as part of the incident response workflow.

Incident Response

From detection to resolution in seconds. The operator approves — ArenaOS executes, logs, and summarizes the impact.

How It Works

From physical reality to digital command

A seamless, real-time pipeline: sensors feed AI, AI drives the twin, the twin empowers decisions.

IoT Sensors
Computer Vision
Telemetry Aggregator
AI Agents
Decision Engine
Digital Twin
Operator UI
<1s
Incident Detection Latency
92%
AI Confidence on Anomaly Detection
42%
Avg. Wait Time Reduction
24/7
Autonomous System Monitoring
Technical Architecture

Built on a modern intelligence stack

Four layers. One coherent operating system.

Perception Layer
OpenCV • Haar Cascades • Live Webcams
Backend Layer
Python • FastAPI • WebSockets
Intelligence Layer
NVIDIA NIM • Llama-3.3-70B • Zustand State
Interface Layer
Next.js • React Three Fiber • Framer Motion
Divyansh Kumar

Divyansh Kumar

B.Tech Artificial Intelligence & Data Science

The Builder

Why I built ArenaOS AI

ArenaOS AI started from a simple question: why are operators still forced to monitor dozens of disconnected screens during critical incidents?

Modern venues generate enormous data from CCTV, sensors, access control, and building systems — but that information is scattered across silos. When a crowd surge or medical event occurs, there is no unified picture of reality.

I built ArenaOS AI to demonstrate how Digital Twins, Computer Vision, and AI agents can collapse that complexity into a single spatial command center — where the software explains the situation and the operator makes the call.

Role
AI & Full Stack Developer
University
MITS Gwalior
Interests
AI Agents, Digital Twins
Focus
Computer Vision, System Design