A cryptocurrency trading and backtesting framework developed specifically for a hackathon hosted on Kaggle.

This project enables participants to build, test, and evaluate algorithmic trading strategies in a structured, extensible environment.

This framework enables teams to:

• Develop algorithmic trading strategies for cryptocurrency markets
• Backtest strategies against historical market data
• Evaluate performance using industry-standard metrics
• Package and submit strategies for evaluation in a standardized format

The project is organized into the following key directories:

• exchange/: core trading engine and backtesting logic
  • engine.py: simulates market conditions and executes trades
• scripts/: utility scripts
  • download.py: used to download and format market data
• strategy/: user-implemented trading strategies
  • main.py: entry point with the on_data function
  • Create your own strategy.py to develop custom strategies

This project uses just for streamlined development workflows.

To get started:

1. Install just
2. Set up the Python environment: just install
3. View available commands: just print

# Set up your Python environment (Python 3.11+ recommended)
just install

# Download default market data (ETH/USDT, BTC/USDT, ETH/BTC)
just download

# Package your strategy for submission
just tar

# Run the backtest and score your strategy
just score

# View current configuration
just print

Strategies will be scored using a combination of profitability, risk-adjusted performance, and trading efficiency.

Participants are required to submit a .tar.gz archive of their strategy directory, including the entry point and any custom logic.

To produce a submission archive in the required format run: just tar