[hnews_account_1]

That is currently a non issue since software is lagging hardware. So you can join 2 systems together fairly inefficiently and still make a go of it. Sure, a good language for people down the stack can topple python from the top, but it won't be easy. It'd be far easier if we were approaching software limits on current hardware and needed to pull everything into one language and system to make it work together.

[yahyaheee]

Its useful for ML in that it requires so much optimization, that being about to drop down in the same language is incredibly handy. For ML software isn't lagging hardware its the other way around.