Aegis-C9 Assistant Coach 🎮🤖

AI-Powered Real-Time Esports Analytics Platform for Cloud9

An intelligent coaching assistant that provides live ML-driven tactical insights for League of Legends and VALORANT matches. Built for the Cloud9 organization to enhance competitive performance through real-time data analysis and predictive modeling.

---

🔄 Hybrid Data Model

Aegis-C9 utilizes a Hybrid Data Architecture that combines historical analysis with real-time intelligence:

📚 Historical Baseline (Kaggle)

• 124,500+ professional match samples from Kaggle datasets
• Used for training XGBoost ML models with comprehensive feature engineering
• Provides statistical baselines for win probability calculations
• Enables pattern recognition across thousands of pro-level games

⚡ Real-Time Ingestion (GRID API)

• GRID Esports Data API integration for live match telemetry
• Real-time player statistics, objectives, and game state updates
• Live event streaming for tactical insight generation
• Seamless fallback to simulation when live data unavailable

This hybrid approach ensures robust predictions grounded in historical data while maintaining real-time responsiveness during live matches.

---

✨ Features

🎯 Real-Time Win Probability

• XGBoost ML model trained on 124,500+ professional match samples
• 91.2% model accuracy with 0.945 ROC-AUC score
• Live probability updates based on game state

📊 ML-Powered Tactical Insights

• Objective Control Analysis (22% weight) - Dragon/Baron priority recommendations
• Gold Efficiency Tracking (18% weight) - Farm pattern optimization
• KDA Ratio Monitoring (15% weight) - Player positioning suggestions
• Damage Efficiency (14% weight) - Target focus recommendations
• Vision Control (12% weight) - Ward placement priorities
• Kill Participation (10% weight) - Team fight involvement analysis
• Survival Rate (9% weight) - Death prevention strategies

🎮 Multi-Game Support

• League of Legends - Full dashboard with champion stats, objectives, and gold tracking
• VALORANT - Round-by-round analysis with agent performance metrics

📈 Live Dashboard

• Player performance cards with real-time KDA, CS, and impact scores
• Win probability chart with historical trend
• Tactical Comms feed with ML-driven suggestions
• Feature importance visualization
• Match configuration settings
• JSON export functionality

---

🛠️ Tech Stack

Frontend

• Next.js 16.1.4 - React framework with App Router
• React 19 - UI components
• TypeScript - Type safety
• Tailwind CSS - Styling
• Recharts - Data visualization
• Lucide React - Icons

Backend

• FastAPI - Python API framework
• XGBoost - Gradient boosting ML models
• scikit-learn - Model training and preprocessing
• joblib - Model serialization
• Uvicorn - ASGI server

Development Tools

• WebStorm - Frontend IDE
• PyCharm - Backend IDE
• Junie - AI-assisted development

---

🚀 Getting Started

Prerequisites

• Node.js 18+
• Python 3.9+
• npm or yarn

Installation

1. Clone the repository

   git clone https://github.com/Princess-B-Kwaniya/Aegis-C9-assistant-coach.git
   cd Aegis-C9-assistant-coach

2. Install frontend dependencies

   npm install

3. Install backend dependencies

   cd aegis_c9_backend
   pip install -r requirements.txt

4. Start the backend server

   cd aegis_c9_backend
   python main.py

   Backend runs on http://localhost:8000

5. Start the frontend (new terminal)

   npm run dev

   Frontend runs on http://localhost:3000

---

📁 Project Structure

Aegis-C9-assistant-coach/
├── aegis_c9_backend/           # Python FastAPI backend
│   ├── main.py                 # API endpoints & ML predictors
│   ├── bridge.py               # GRID API integration
│   ├── find_live_match.py      # Live match detection
│   └── data/
│       ├── lol/                # LoL ML model & training
│       │   ├── lol_model.json
│       │   ├── scaler.joblib
│       │   └── train_lol_model.py
│       └── valorant/           # VALORANT ML model
│           ├── valorant_model.json
│           └── scaler.joblib
├── src/
│   ├── app/                    # Next.js pages
│   │   ├── page.tsx            # LoL dashboard
│   │   └── valorant/page.tsx   # VALORANT dashboard
│   ├── components/
│   │   ├── dashboard/          # Dashboard components
│   │   ├── layout/             # Sidebar & navigation
│   │   └── modals/             # Settings modal
│   ├── hooks/                  # React hooks
│   │   ├── useAegisLive.ts     # LoL data hook
│   │   └── useValorantLive.ts  # VALORANT data hook
│   └── types/                  # TypeScript definitions
└── public/                     # Static assets

---

🔌 API Endpoints

Endpoint	Method	Description
/	GET	Health check
/lol-predictions	GET	LoL ML predictions & game state
/valorant-predictions	GET	VALORANT ML predictions
/api/stats	GET	Raw match statistics
/stream-telemetry	GET	Live telemetry stream (SSE)

---

🧠 ML Model Details

LoL Predictor (XGBoost)

• Training Data: 124,500 professional match samples from Kaggle LoL Esports Dataset
• Features: 15 engineered features from raw game stats
• Target: Win/Loss binary classification
• Performance: 91.2% accuracy, 0.945 ROC-AUC
• Real-Time Updates: GRID API provides live game state for prediction refinement

Feature Engineering

# Key features used by the model
- kda_ratio = (kills + assists) / (deaths + 1)
- gold_efficiency = gold_earned / duration_mins
- damage_efficiency = damage_to_champ / damage_dealt
- vision_per_min = vision_score / duration_mins
- objective_control = (barons*3 + dragons*2 + towers) / 10

---

🎨 Screenshots

Dashboard showing live win probability, player stats, and ML tactical insights

---

👥 Team

Built with ❤️ for Cloud9

---

📄 License

This project is for educational and demonstration purposes.

---

� Data Sources

Source	Purpose	Data Type
Kaggle	Historical baseline & ML training	124,500+ pro match records
GRID API	Real-time match ingestion	Live telemetry & events

---

🙏 Acknowledgments

• Cloud9 Esports Organization - Inspiration and use case
• Kaggle - Historical esports match datasets for ML training
• GRID Esports Data API - Real-time match data integration
• JetBrains - WebStorm, PyCharm, and Junie AI development tools