Our take
- Activity
- 53
- Community
- 92
- Issues
- 72
- Pull requests
- 78
+155 stars in the last 10 days
- Merges more pull requests than 83% of the projects we track
- More contributors than 77% of the projects we track
Overview
Lean is an event-driven algorithmic trading engine for building and running strategies across financial markets. It provides a shared platform for researching ideas, backtesting them against market data, and deploying algorithms for live trading.
The engine is designed for quantitative developers who need customizable trading components and support for both C# and Python. Its command-line interface can manage projects and run local research, backtests, optimization, and live trading workflows.
Key Features
- Event-driven engine for algorithmic trading.
- Supports C# and Python algorithm development.
- Provides backtesting and live trading workflows.
- Modular architecture with customizable components.
- Supports alternative data and multiple financial markets.
- LEAN CLI commands for project creation, research, backtesting, optimization, and live trading.
- Local development with Docker-based CLI workflows and cloud-assisted development options.
Use Cases
- Quant developers can evaluate a trading strategy locally before considering live deployment.
- C# or Python developers can build systematic strategies using a shared trading engine.
- Researchers can use the CLI's local Jupyter environment to explore algorithm ideas.
- Teams can customize engine components to fit specific data or trading workflows.
Getting Started
Install the CLI with pip install lean, then use lean project-create to create a starter project. See the README for setup details and platform-specific instructions.
Alternatives
- FinRL: FinRL focuses on reinforcement-learning trading workflows, while Lean is a general event-driven engine for backtesting and live strategies.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| Lean | C# | Apache-2.0 | 21.9k | Not checked yet |
| FinRL | Jupyter Notebook | MIT | 16.5k | Active |
Considerations
Running the engine locally requires the .NET SDK and a suitable development setup. The CLI's research, backtesting, optimization, and live workflows use Docker. Live trading involves additional setup beyond installing the CLI. Lean is a flexible engine rather than a ready-made strategy, so users need to supply their own algorithms and assess their risks.
Found this useful?
Share it with someone who would like Lean.