Hi everyone! My name is Damien. I'm a Data Scientist and I'm quite familiar with luigi https://github.com/spotify/luigi (I contributed once to fix a tiny bug). prefect seems fun and reliable. I'll try it. Maybe I would have some questions

I see alot of people comparing Prefect with Airflow, but can someone give me a few points how Prefect is different from Luigi? They both use Python as API and use Task definitions and dependencies

It’s been a while, but IIRC Luigi uses a highly-stateful notion of “targets” to determine whether a task finished successfully. These targets conflate data outputs and state outputs, meaning you can’t adjust the state of a task without modifying the data it produced. Conversely, Airflow only supports state outputs. Prefect supports both state and data outputs independently.

Thanks @Jeremiah 🙂 I would not have better explained