I’ve given a very introductory talk about what Arrow “gives for free” when using the right kind of UDF. It’s more fun in person, but with the references at the end and the presenter notes I think you could get an idea of what you will want to mention quicker than having to look at it from scratch. It’s [here](https://github.com/rberenguel/pyspark-arrow-pandas), I hope you find it useful!