PandaAI Quantflow is an integrated quantitative trading and machine learning workflow platform designed for quantitative researchers, aiming to democratize strategy development. It offers an end-to-end solution for tasks ranging from data processing and feature engineering to backtesting and algorithmic trading, lowering the barrier to entry for AI-driven finance.

How It Works

The platform utilizes a node-based visual workflow designer for building complex quantitative research and trading strategies. It integrates mainstream machine learning algorithms (XGBoost, LightGBM, etc.) and provides a high-performance, event-driven backtesting engine compatible with the PandaFactor analysis framework. A microservices architecture with a FastAPI backend supports distributed task execution, multi-user management, and custom plugin development via a @work_node decorator.

Quick Start & Requirements

• Installation: pip install -e . within the cloned repository.
• Prerequisites: Anaconda environment, PandaAI-provided database (download required), and panda_factor dependencies. MongoDB must be running with authentication enabled.
• Running: Start the server with python src/panda_server/main.py. Access UI at http://127.0.0.1:8000/charts/ and http://127.0.0.1:8000/quantflow/.
• Docs: PandaFactor

Highlighted Details

• Visual workflow designer with drag-and-drop functionality.
• Supports custom node development via a plugin system.
• High-performance backtesting engine for stocks and futures.
• Enterprise-grade service architecture with REST API and distributed execution.

Maintenance & Community

• Open to contributions via Issues and Pull Requests.
• Community support via group chat (details in README).

Licensing & Compatibility

• Licensed under GPLv3. This copyleft license may require derivative works to also be open-sourced under GPLv3, potentially impacting commercial use or integration into closed-source projects.

Limitations & Caveats

The project requires a local MongoDB instance with specific data and authentication setup, which can be time-consuming. MacOS installation is listed as "in preparation." Full production readiness and specific performance benchmarks are not detailed.