Trading System Documentation

A Production-Grade Algorithmic Trading System

Built with Python, PostgreSQL, Redis, and Modern Web Technologies

Technology Stack

Python PostgreSQL Redis FastAPI pandas

🚀 Quick Start

Get your trading system up and running in minutes:

# Clone and setup
git clone https://github.com/nishantnayar/trading-system.git
cd trading-system

# Activate your existing conda environment
conda activate your-environment-name

# Install dependencies
pip install -r requirements.txt

# Configure and run
cp deployment/env.example .env
python scripts/setup_databases.py
python main.py

Access your Streamlit UI: http://localhost:8501

📋 What You'll Find Here

Getting Started

User Guide

API Reference

Development

🏗️ System Architecture

graph LR
    A[Data Sources] --> B[Data Ingestion]
    B --> C[Core Services]
    C --> D[Frontend]
    B --> E[PostgreSQL]
    C --> E

The Trading System follows a modular, service-oriented architecture:

Data Flow: - Data Sources → Polygon.io, Yahoo Finance, Alpaca API - Data Ingestion Service → Processes and stores market data - Core Services → Strategy Engine, Risk Management, Execution, Analytics - Storage → PostgreSQL (data persistence) and Redis (caching) - Orchestration → Prefect workflows for scheduled tasks - Frontend → Streamlit UI and FastAPI REST API

Service Interactions: - Data Ingestion feeds data to Core Services - Strategy Engine → Risk Management → Execution pipeline - All services store data in PostgreSQL - Analytics service provides metrics to the UI - Prefect orchestrates scheduled data ingestion and analytics workflows

For detailed architecture documentation, see System Architecture.

Core Components

Component Technology Status Purpose
Data Ingestion Python + pandas ✅ v1.0.0 Market data processing (Polygon.io, Yahoo Finance, Alpaca)
Strategy Engine Python + Pydantic 🚧 v1.1.0 Algorithm execution (planned)
Risk Management Python + PostgreSQL 🚧 v1.1.0 Risk controls (planned)
Execution Engine Python + Alpaca API ✅ v1.0.0 Order management (account, positions, orders)
Analytics Python + Plotly ✅ v1.0.0 Performance tracking and visualization
AI Services Python + Ollama ✅ v1.0.0 Natural language processing for stock screener
Database PostgreSQL ✅ v1.0.0 Data persistence
Cache Redis ✅ v1.0.0 High-speed access (optional)

🎯 Key Features

Trading Capabilities

  • ✅ Paper Trading - Start with Alpaca paper trading (v1.0.0)
  • ✅ Historical Data - Polygon.io and Yahoo Finance integration for backtesting (v1.0.0)
  • ✅ AI-Powered Screener - Natural language stock screening with Ollama LLM (v1.0.0)
  • 🚧 Strategy Backtesting - Test before live deployment (v1.1.0)
  • 🚧 Risk Controls - Built-in risk management (v1.1.0)
  • ✅ Performance Analytics - Interactive charts and metrics with Plotly (v1.0.0)
  • ✅ Order Management - Account, position, and order viewing/cancellation (v1.0.0)
  • 🚧 Order Placement - Place new orders via UI (v1.1.0)

Technical Features

  • ✅ Modular Architecture - Service-oriented design ready for scaling
  • ✅ Type Safety - Pydantic for data validation
  • ✅ Data Processing - pandas for analytics and manipulation
  • ✅ Modern UI - Streamlit + Plotly Charts + Custom CSS
  • ✅ Professional Charts - Interactive financial visualizations with Plotly
  • ✅ AI Integration - Local LLM support via Ollama for natural language queries
  • ✅ Session State - Persistent data sharing across pages
  • ✅ Comprehensive Logging - Database-first logging with PostgreSQL storage and file fallback
  • ✅ Database Design - PostgreSQL with comprehensive schema
  • ✅ Data Quality - Automated validation and monitoring
  • ✅ Timezone Support - UTC storage with Central Time display

Development Features

  • ✅ Code Quality - Flake8, Black, isort, mypy
  • ✅ Documentation - MkDocs
  • ✅ Testing - Comprehensive test coverage
  • ✅ CI/CD - Automated quality checks

📊 Dashboard Preview

Your Streamlit trading interface provides:

  • Portfolio Management - Real-time portfolio tracking and performance metrics
  • Market Analysis - Interactive charts with technical indicators using Plotly
  • Stock Screener - AI-powered stock screening with natural language queries (Ollama integration)
  • System Information - Team details and system architecture
  • Settings - User preferences and session state management
  • Session State - Persistent data sharing across all pages
  • Modern UI - Clean, professional Streamlit interface
  • Timezone Support - All data displayed in Central Time (stored in UTC)

🔧 Configuration

Environment Variables

# Database
POSTGRES_URL=postgresql://user:pass@localhost:5432/trading
REDIS_URL=redis://localhost:6379/0

# Alpaca API
ALPACA_API_KEY=your_api_key
ALPACA_SECRET_KEY=your_secret_key
ALPACA_BASE_URL=https://paper-api.alpaca.markets

# Ollama (Optional, for AI features)
OLLAMA_BASE_URL=http://localhost:11434

# Logging (Database-first with file fallback)
LOG_LEVEL=INFO
# Logs are stored in PostgreSQL logging schema by default
# See config/logging.yaml for detailed configuration

Service Ports

  • Streamlit UI: http://localhost:8501
  • API Docs: http://localhost:8001/docs
  • PostgreSQL: localhost:5432
  • Redis: localhost:6379

🚨 Important Notes

Paper Trading First

  • Start with Alpaca paper trading
  • Test your strategies thoroughly
  • Understand risk management
  • Only move to live trading when ready

Risk Management

  • Set appropriate position sizes
  • Use stop-losses and take-profits
  • Monitor your exposure
  • Never risk more than you can afford to lose

Development

  • Follow the coding standards
  • Write comprehensive tests
  • Document your changes
  • Use the logging system effectively

📞 Support

Documentation

Getting Help

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

Built by Nishant Nayar

GitHub Documentation Issues