A premium, real-time trading intelligence platform with AI-powered market predictions
Real-time market monitoring with portfolio summary, live candlestick charts, market news feed, and active alerts panel.
Advanced technical analysis with price charts, SMA/RSI indicators, and AI-driven sentiment analysis.
AI-powered stock predictions with LSTM neural network confidence scores and one-click trading.
TradingLab is a professional-grade trading observability platform that combines real-time market data, technical analysis, and AI-powered LSTM predictions to provide traders with actionable intelligence. Built with a modern microservices architecture, it seamlessly integrates Next.js for the frontend, Node.js for real-time data streaming, and a dedicated Python ML service for deep learning predictions.
- AI Prediction Visualizer: Real-time confidence scores (0-100) with visual prediction bars
- Live Market Data: WebSocket-powered price updates with sub-second latency
- Smart Trading Interface: One-click Buy/Sell execution with position tracking
- Portfolio Analytics: Real-time P&L, gainers/losers, and active stock monitoring
- LSTM Neural Network: PyTorch-based deep learning model for price prediction
- Automated Training Pipeline: Historical data ingestion from Alpha Vantage
- Confidence Scoring: Probabilistic predictions with uncertainty quantification
- Microservice Architecture: Dedicated Python FastAPI service for ML workloads
- Multi-Indicator Analysis: RSI, MACD, SMA, Bollinger Bands
- Health Score Algorithm: Composite scoring based on technical signals
- Interactive Charts: D3.js candlestick charts with multiple timeframes
- Volume & P/E Integration: Fundamental data alongside technical indicators
- Position Management: Track buy/sell orders with entry prices and timestamps
- Trade Execution Dialog: Quantity input with real-time total calculation
- Portfolio Persistence: JSON-based storage for trade history
- Real-time Notifications: Toast alerts for successful trades
- News Aggregation: Real-time financial news via Alpha Vantage
- Sentiment Analysis: Automated bullish/bearish scoring
- MCP Integration: Model Context Protocol for news streaming
- Advanced Filtering: Search and filter by ticker, sentiment, or date
- Price Triggers: Set "Above" or "Below" alerts for any ticker
- Live Notifications: Instant dashboard notifications via WebSocket
- Make.com Integration: Webhook support for external automation
- Dark/Light Themes: System-aware theme switching
- Responsive Design: Optimized for desktop, tablet, and mobile
- Profile Management: Customizable user profiles with avatars
- Collapsible Sidebar: Adaptive navigation for maximum screen space
┌─────────────────────────────────────────────────────────────┐
│ TradingLab │
├─────────────────────────────────────────────────────────────┤
│ │
│ ┌──────────────┐ ┌──────────────┐ ┌──────────┐ │
│ │ Frontend │◄────►│ Backend │◄────►│ ML Service│ │
│ │ (Next.js) │ │ (Node.js) │ │ (Python) │ │
│ │ Port: 3000 │ │ Port: 3001 │ │Port: 8080│ │
│ └──────────────┘ └──────────────┘ └──────────┘ │
│ │ │ │ │
│ │ │ │ │
│ ▼ ▼ ▼ │
│ ┌──────────────────────────────────────────────────────┐ │
│ │ Alpha Vantage API │ │
│ │ (Market Data, News, Technical Indicators) │ │
│ └──────────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────┘
- Framework: Next.js 14 (App Router)
- Styling: Tailwind CSS 4
- UI Components: shadcn/ui
- Charts: D3.js
- State Management: React Hooks
- WebSocket: Socket.io-client
- Notifications: Sonner
- Runtime: Node.js 18+
- WebSocket Server: Socket.io
- API Integration: Alpha Vantage SDK
- MCP Client: Model Context Protocol
- Caching: In-memory with TTL
- Framework: FastAPI
- Deep Learning: PyTorch 2.0
- Data Processing: Pandas, NumPy
- Preprocessing: scikit-learn
- Model: LSTM (Long Short-Term Memory)
- Containerization: Docker & Docker Compose
- Orchestration: Multi-stage builds
- Networking: Internal Docker network
- Volumes: Hot-reloading for development
- Docker & Docker Compose (recommended)
- Docker Desktop for Windows/Mac
- Docker Engine + Docker Compose for Linux
- Alpha Vantage API Key (free tier available)
- Get your key: https://www.alphavantage.co/support/#api-key
-
Clone the repository
git clone https://github.com/yourusername/trading-observer.git cd trading-observer -
Configure environment variables
# Create .env file in the root directory echo "ALPHA_VANTAGE_API_KEY=your_api_key_here" > .env echo "WS_PORT=3001" >> .env
-
Start all services
docker compose up -d --build
-
Access the application
- Frontend: http://localhost:3000
- Backend WebSocket: http://localhost:3001
- ML Service: http://localhost:8080
# Install dependencies
npm install
# Start the WebSocket server
npm run server
# In a new terminal, start the frontend
npm run dev# Navigate to ML service directory
cd src/ml-service
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
# Start the ML service
uvicorn main:app --host 0.0.0.0 --port 8000trading-observer/
├── src/
│ ├── app/ # Next.js App Router
│ │ ├── (dashboard)/ # Dashboard routes
│ │ │ ├── watchlist/ # StockPredict page
│ │ │ ├── market/ # Market analysis
│ │ │ ├── forecasting/ # AI predictions
│ │ │ ├── alerts/ # Alert management
│ │ │ └── news/ # News feed
│ │ └── api/ # API routes
│ │ ├── forecasting/ # Prediction endpoint
│ │ ├── market/ # Market data
│ │ └── positions/ # Trade management
│ ├── components/
│ │ ├── layout/ # Dashboard layout
│ │ ├── watchlist/ # StockCard, TradeDialog
│ │ ├── market/ # Charts, analysis
│ │ ├── providers/ # Context providers
│ │ └── ui/ # shadcn components
│ ├── services/
│ │ ├── market-service.ts # Alpha Vantage client
│ │ └── forecasting-service.ts # Technical analysis
│ └── ml-service/ # Python ML Service
│ ├── main.py # FastAPI app
│ ├── model.py # LSTM architecture
│ ├── train.py # Training pipeline
│ ├── data_loader.py # Data fetching
│ ├── requirements.txt # Python dependencies
│ └── Dockerfile # ML service container
├── server/
│ ├── index.js # WebSocket server
│ └── mcp-client.js # MCP integration
├── docker-compose.yml # Multi-service orchestration
├── Dockerfile # Frontend/Backend container
└── README.md # This file
# Access the ML service container
docker exec -it trading-observer-ml bash
# Train the model (example: Apple Inc.)
python train.py --symbol AAPL
# The trained model will be saved to:
# artifacts/AAPL_model.pth
# artifacts/AAPL_scaler.pklThe LSTM model uses the following hyperparameters (configurable in train.py):
- Sequence Length: 60 days (look-back window)
- Hidden Dimensions: 50
- LSTM Layers: 2
- Epochs: 20
- Batch Size: 32
- Learning Rate: 0.001
- Train/Test Split: 80/20
Input (60 days of price data)
↓
LSTM Layer 1 (50 hidden units)
↓
LSTM Layer 2 (50 hidden units)
↓
Fully Connected Layer
↓
Output (Next day price prediction)Returns AI prediction and technical analysis for a stock.
Response:
{
"symbol": "AAPL",
"healthScore": 72,
"recommendation": "BUY",
"signals": {
"rsi": "neutral",
"macd": "bullish",
"sma": "bullish"
},
"technicalData": {
"rsi": 55.2,
"macd": 1.23,
"sma": 188.5,
"price": 189.17,
"volume": "52340000",
"peRatio": 28.5
}
}Creates a new trade position.
Request:
{
"symbol": "AAPL",
"type": "buy",
"quantity": 10,
"entryPrice": 189.17
}Health check endpoint.
Response:
{
"status": "healthy"
}Request prediction for a stock.
Request:
{
"symbol": "AAPL",
"days": 1
}| Service | Port | Description |
|---|---|---|
frontend |
3000 | Next.js application |
backend |
3001 | WebSocket server |
ml-service |
8080 | Python ML API |
./src/ml-service:/app- Hot-reloading for ML service development./positions.json:/app/positions.json- Trade persistence
Create a .env file in the root directory:
# Required
ALPHA_VANTAGE_API_KEY=your_api_key_here
# Optional
WS_PORT=3001
PORT=8000 # ML service internal port- Real-time market data streaming
- Technical analysis indicators
- StockPredict dashboard
- Trading functionality
- ML service infrastructure
- LSTM model architecture
- Training pipeline
- Connect ML predictions to frontend
- Real-time inference API
- Model versioning and A/B testing
- Multi-asset portfolio optimization
- Backtesting engine
- Risk management tools
- Social trading features
- Mobile app (React Native)
- Multi-user support
- Role-based access control
- Advanced charting (TradingView integration)
- Broker API integration (Alpaca, Interactive Brokers)
- Cloud deployment (AWS/GCP)
Contributions are welcome! Please follow these steps:
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
This project is licensed under the MIT License - see the LICENSE file for details.
- Alpha Vantage for market data API
- shadcn/ui for beautiful UI components
- PyTorch team for the deep learning framework
- Next.js team for the amazing React framework
For questions or support, please open an issue on GitHub.
Built with ❤️ by the TradingLab Team
⭐ Star this repo if you find it useful!