Is PySide2 actually ready from primetime use? The list of known issues on the wiki doesn't fill me with confidence and there are a large number of bugs on the tracker.
I'm in agreement that Qt for Python (PySide2) is likely the best option in the long term due to the first party maintenance by Qt, but I'm concerned it's not as stable as PyQt5 yet.
I haven't noticed anything major as of yet, and anyway, you know how it goes: adoption usually speeds up bugfixes.
I'd recommend to start building with PySide2; by time your project is finished, it will be stable. Or else, worse case, you just change your import statements and leave the rest the same. ;)
I already have a large project that was ported from PySide1 to PyQt and it was non-trivial to do (sure most of it was the same but that couple of percent difference was a pain to find and update), so I'm not keen to repeat that experience until it's worth it. Sounds like maybe I should wait a bit longer or look into using qtpy as an abstraction layer.
I think that was a newer thing about Qt5...that PySide2 and PyQt5 had identical APIs. But I could be wrong. The abstraction layer sounds wise in any case.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.