Hi Qlib team,

First of all, thanks for building such a strong open-source quant research platform. Qlib’s pipeline from data processing, modeling, backtesting to portfolio research has been very inspiring, especially for people working on AI-oriented quant workflows.

I am building a related but narrower A-share research project: Factor Lab. It focuses on China A-share factor screening, profit-gap / event-driven stock pools, strategy review, and AI-generated research reports. Instead of trying to replace a full quant platform like Qlib, Factor Lab is more like a productized research workstation around one specific workflow: discover candidates, score them with factors, then let multiple AI roles debate the stock and generate an auditable report.

I would love to exchange ideas around:

• how to connect a vertical A-share factor stock pool with Qlib-style model/backtest pipelines;
• whether AI research reports can be used as an additional layer for factor validation;
• how to track historical stock pool decisions so the research process is reproducible;
• whether event-driven signals such as profit-gap candidates can be modeled cleanly inside a broader quant framework.

Project link for reference: https://www.afactorlab.com/

Happy to discuss if this direction is interesting to the Qlib community.