Hi HN, I built Ciaren, an open-source visual workflow tool for Python data pipelines.

You can build a pipeline visually, run it locally, and export the resulting workflow as real Python/Polars/Pandas code.

I built it because I wanted the convenience of a visual workflow without sacrificing readable, portable code or relying on a proprietary runtime.

I'm particularly interested in feedback from Python/data engineers: does this abstraction actually make sense, or is it more useful to just write the Polars code directly?

GitHub: https://github.com/ciaren-labs/Ciaren