Joel

Thanks - @npfp please do! It will be a great place to co-design the best solution possible 💪

Hi @crash4 I completely get your reasoning here and also like your solution. In general parallelism in Python can be a pain and my fear is that it would be...

I think the big issue is when you transcode between Spark, Pandas, Polars etc. I like this default behaviour, we can throw a warning if people really feel strongly.

The value in shareable Kedro Viz is often about sharing updates with business stakeholders. The tutorial is not necessary in this context. It would be helpful if this was disabled...

I would like URL flags personally, that would be easiest thing to send stakeholders as you're already going to send a URL to them.

Heavy dependency imports would be great here too

Sorry I meant things like Pytorch / Tensorflow / Spark / Pandas

I also have evidence from the internal 'managed analytics' days (serving pipeliens as typesafe FastAPI apps) that Session was too heavy in terms of start up time. The implementation bypassed...

>Anything that I'm missing? Only the fact that advanced users will likely patch things like the context, sessions and custom CLIs which make it near impossible to upgrade. I still...

I was remaking with the team a couple of weeks ago it feels like 1.0.0 is in sight for the first time, I would argue earnestly look at where we...