How to embed vomit in your own code, by JetBrains

https://blog.jetbrains.com/research/2026/07/kotlinllm-open-source/

You can now embed prompts in your Kotlin code. It's worse than generating code and checking it beforehand since it will now be done at runtime. Expect a lot of new applications that will put one prompt in the main function like:

fun main() { chatgpt("open a window with a button, and make no mistakes") }  
112 points · 13 comments · view on lemmy.world

13 Comments

eleijeep@piefed.social · 31 pts · 18d (1 reply)

JetBrains? More like ShitForBrains amirite lmao gotem

snoons@lemmy.ca · 5 pts · 18d

Well, some of the data centres are burning cow shit for power so...

Yaky@slrpnk.net · 28 pts · 18d (1 reply)

(Poe's law activate)

Is this a joke or trolling? This is an RNG for crashing the app with a runtime error.

one_old_coder@piefed.social · 10 pts · 18d

Not a joke I guess. They have developed their own "LLM proxy," their own agent, and now they put prompts in the code.

addie@feddit.uk · 27 pts · 18d (2 replies)

If JetBrains could spend more time fixing all their broken shit and less time on this, that would be amazing.

Had my copy of IntelliJ spontaneously deactivate itself yesterday, presumably a network glitch contacting my company's auth servers. Second attempt successful, and enabled every fucking 'premium' plugin while it was at it. Took half a day to turn all that shit back off again and find every setting that was changed before I could do local builds again. Aargh.

Main selling point of IDEA over eg. Eclipse is that it's "faster". Getting slower all the time, and I've never had Eclipse spontaneously shit itself in quite such fashion.

lichtmetzger@discuss.tchncs.de · 4 pts · 17d

If JetBrains could spend more time fixing all their broken shit and less time on this, that would be amazing.

After each update of WebStorm I have to manually edit the VM options to enable Vulkan support, otherwise it will render the editor in software mode with ~5fps.

Smooth-scrolling a little bit of text in 2026 seems way too difficult for a commercial project. If AI is so great, why can it not implement that?

quick_snail@feddit.nl · 3 pts · 17d

If they're using AI, it's probably going to get worse, not better.

Switch to something that has a no-AI policy

neclimdul@lemmy.world · 24 pts · 18d (2 replies)

This is brilliant. Just think about your bug reports.

"Doesn't work. Saw XYZ"

/close Could not reproduce.

Dojan@pawb.social · 12 pts · 18d

Restart application until it works.

Bananas.

one_old_coder@piefed.social · 3 pts · 18d

That's a great idea. Once you know there is random stuff happening in the code, you can close every bug report!

fargeol@lemmy.world · 14 pts · 18d (1 reply)

That's suboptimal... In Rust we do that at compile-time: https://github.com/germangb/ai-bindgen

massive_bereavement@fedia.io · 5 pts · 18d

What could go wrong.

lichtmetzger@discuss.tchncs.de · 2 pts · 17d

JetBrains is just weird. Their plugin for WebStorm technically supports local LLMs, but it defaults to OpenAI models and after each update I have to reconfigure it again to use my own machine instead of a massive datacenter poisoning freshwater.

One time it had a hiccup and couldn't process my input and when I clicked a simple "Resend" button, it just randomly decided to send all of my code into the cloud to be handled by OpenAI instead of my local machine. Thanks, that's totally what I want you fucking morons.

Cloud providers cannot be disabled in the plugin and the "Automatic" setting in the model selection does not mean it'll always use your own local LLM, it just does what it wants.

Their priorities seem pretty clear here - they added support for local LLMs because otherwise they'd piss off legitimate developers, but it's very barebones and the actual focus is on stupid cloud LLM providers like Anthropic and OpenAI. Can't wait for this bubble to burst.