I’ve gained quite an expertise in BP and I’ve made quite a complex game that at one point was 98%+ written in BP. Despite that, I support deprecation of BPs - as long as Verse is a proper replacement.
In this video below…
https://www.youtube.com/watch?v=JvLpB_ZQHgg
…my game is in a state where it runs almost completely on BPs. It’s very performant and well organized (BPs are neither slow nor spaghetti when used correctly).
But even after gaining and utilizing so much BP experience, I believe Blueprints are bit of a trap. They are super easy to get started with, but as soon as things start to get more complex, you pay a LOT in inability to refactor, to do large scale changes and risk corruption issues that are often irreversible (reverting version control commit aside), since the format is binary and non human readable.
At some point, changing large system feel discouraging because of how many BP classes the changes ripple across, meaning lots, I mean LOTs of manual, error prone rewiring and copying.
Another big obstacle of BPs is that they, up until recently could not be read, written, edited and refactored by AI LLM models, which are increasingly used by anyone who’s not a delusional activist, and has to do the job for actual living, not just as a hobby.
Even with the use of MCP servers, the AI models will have to parse a lot more metadata about the language than they should… such as node positions, connections, socket states, etc… They will have to do much, much more work to write BP code than they’d have to do to write Verse code.
Due to these issues, I had to rewrite most of my game to Angelscript, but I am still not happy, because maintaining separate unofficial, source-built fork of UE for a solo/indie project creates a lot of overhead/friction and causes a lot of issues that waste time.
So my hope is for Verse to finally be a proper scripting language for UE that does not require editor restart for every minor change the way C++ does. C++ is a truly abysmal language that takes all the fun out of creation, and you spend more time solving the problems of the language than you spend solving the problems of your game.
From my experience, when I go from BP to C++, I usually go from ~10 new features per day to about ~1 new feature per 10 days. And Angelscript/C# experience has proven to me it’s not because of node based vs text based, but because C++ is a true outlier in terms of how terrible and unergonomic a programming language can be.
UE has been in a dire need for intermediate level, C# like scripting language for ages.
My only worry is Verse UX. Epic has proven time and time again, that unless it’s a core engine feature, no one gives a crap about UX. Epic Launcher, FAB, this forum, learning portal, they are proving it time and time again.
Verse is currently in a really bad place not because of the language but because of the tooling around it. For example Verse extension for VSCode repeatedly breaks users’ VSCode theme replacing the color scheme with it’s own, ugly one. It essentially behaves lime a malware and it’s not been fixed for years.
So on the one hand, I am all for replacing BPs with Verse, as it’s appropriate language for the future we’re heading into, but at the same time, I am really worried the Verse ecosystem will get the exact same UX/usability treatment Epic Launcher, FAB, this forum and the learning portal did.
Therefore, my conclusion is: 
