AI-Powered Smart Stadium & Tournament Operations Platform
🚀 Live Demo: https://arena-mind-olive.vercel.app
📂 Repository: https://github.com/Sandesh2311/ArenaMind
ArenaMind AI is a premium-grade GenAI experience for smart stadium operations, designed for PromptWars Virtual 2026 Main Challenge 4. The platform demonstrates how a role-aware, multilingual assistant can support fans, organizers, volunteers, and on-ground staff during high-pressure live events.
ArenaMind AI brings together a modern React frontend, a hybrid AI service layer, and simulated venue operations data to deliver intelligent decision support in a stadium environment. The experience is optimized for rapid demo execution, accessibility, and explainable AI behavior while remaining technically credible for larger-scale deployment planning.
Modern stadium operations depend on fast, coordinated action across multiple roles during peak events. Operators need to balance crowd flow, queue pressure, incidents, communications, and safety decisions in real time. ArenaMind AI addresses this challenge by presenting role-specific dashboards and a guided AI assistant that can answer common requests locally, escalate complex reasoning to Gemini when available, and fall back safely when the model is unavailable.
| View | Preview |
|---|---|
| Fan Dashboard |  |
| Organizer Dashboard |  |
| Volunteer Dashboard |  |
| Experience | Preview |
|---|---|
| AI Assistant |  |
| Crowd Analytics |  |
| Indoor Navigation |  |
| Emergency Dashboard |  |
ArenaMind follows a modular, layered architecture built around React components, reusable hooks, service orchestration, validation utilities, and simulation data. This structure keeps the experience readable, testable, and easy to extend for a future production backend.
The system is designed to support a single-page, low-latency experience with deterministic behavior for common prompts and graceful degradation for model or network failures. It prioritizes clarity over unnecessary complexity, which is important for a hackathon submission that must be understandable to judges, stakeholders, and future engineering teams.
The AI layer uses a hybrid approach:
- Local response matching handles high-confidence, low-risk stadium questions quickly.
- Gemini is invoked only for complex prompts when a valid API key is present.
- Fallback logic ensures that users still receive operationally useful responses during offline or degraded conditions.
- Validation and prompt hardening protect the experience from malformed or injection-like input.
src/
components/ Shared UI, landing page, and dashboards
components/dashboards/organizer/
Organizer-specific analytics sections
constants/ App-wide configuration and assistant context
data/ Venue simulation datasets and content
hooks/ React hooks for assistant and preferences
services/ AI orchestration and fallback services
styles/ Tailwind and global styling entry points
test/ Shared test setup
utils/ Validation, caching, and pure helpers
- React 18 + Vite
- Tailwind CSS
- Recharts for analytics visuals
- Lucide icons
- Vitest + React Testing Library + V8 coverage
- ESLint 9 with React hooks rules
| Capability | Fan | Organizer | Volunteer | Staff |
|---|---|---|---|---|
| Role-specific dashboard | Yes | Yes | Yes | Partial |
| AI assistant with multilingual support | Yes | Yes | Yes | Yes |
| Queue and crowd guidance | Yes | Yes | Yes | Yes |
| Navigation and route cards | Yes | Partial | Yes | Yes |
| Incident and alert monitoring | Partial | Yes | Yes | Yes |
| Task prioritization and coordination | Partial | Yes | Yes | Yes |
| Offline-safe fallback guidance | Yes | Yes | Yes | Yes |
| PromptWars requirement | ArenaMind implementation |
|---|---|
| Smart Stadiums | Unified venue operations experience with role-aware dashboards and operational decision support |
| Tournament Operations | Organizer overview cards, incident summaries, queue analytics, and volunteer monitoring |
| GenAI-enabled Architecture | Hybrid local/Gemini AI flow with cache, validation, and deterministic fallback |
| Crowd Management | Heatmap, queue analytics, high-risk recommendations, and crowd guidance |
| Indoor Navigation | Route cards, gate guidance, alternate path suggestions, and emergency routing |
| Real-time Decision Support | AI-generated next actions, operational insights, and live alert summaries |
| Multi-language Assistance | English, Hindi, Spanish, French, Portuguese, and Arabic support |
| Fans | Fan dashboard for services, seating, parking, food, and route planning |
| Organizers | Organizer dashboard for queue and incident oversight |
| Volunteers | Volunteer dashboard for coordination, routing, and translation help |
| On-ground Staff | Staff-facing operational direction and emergency guidance |
ArenaMind is designed for deployment as a Vite-based static web application. The recommended production target is Vercel, with environment variables configured for Gemini access.
The project includes a full automated testing workflow covering UI interaction, hooks, validation, AI fallback behavior, and dashboard switching.
Verified results from the current repository state:
- 58 tests passing
- 100% statement coverage
- 98.87% branch coverage
The interface is documented and implemented with keyboard support, semantic landmarks, labeled controls, live-region announcements, and accessible chart alternatives.
ArenaMind follows a defensive-by-default approach for prompt handling, input validation, safe fallback behavior, and environment variable usage. Secrets remain out of the browser and client-side storage.
The application is optimized for responsiveness by keeping common prompts local, caching repeated AI responses, minimizing blocking model calls, and using lightweight, component-based rendering.
- Live venue telemetry integration
- Real indoor positioning and geofencing
- Authentication and role-based access control
- Audit logging and incident review workflows
- Expanded multilingual and voice-assisted interaction
Developed by
Sandesh Mishra
PromptWars Virtual 2026
Google for Developers × Hack2Skill
This project is intended for educational, demo, and hackathon evaluation use. Please review local deployment and usage policies before production adoption.