Konstantin

I enjoy solving problems in Laravel, NVIM, and LLMs

Tech Lead @ InnoBrain

[Konstantin @kauffinger]

@steveruizok Wrong, it should be a loop obviously

[Konstantin @kauffinger]

@dexhorthy I use recursive self looping. I tell my main loop in ultracode mode they aren’t allowed to commit and have to spawn an agent who does, but they have to tell them the same thing.

Only the one agent who is sure this is the best commit ever will not listen…

[simp 4 satoshi @iamgingertrash]

If you believe in the singularity,
You are exit liquidity btw

Have fun

[Konstantin @kauffinger]

Your perception of LLM reliability is shaped by:
- how well what you are doing is represented in training data
- how deep your (technical) expertise in that domain is
- how much you care about the system staying maintainable
- how you drive your LLM: /goal or scoped chunks of work, giving it the tools it needs to get real world feedback

[Konstantin @kauffinger]

While it’s true that LLMs can do basically all coding, they definitely do hallucinate. The analogy isn’t „the car you have doesn't exist“, it’s „cars are still an unreliable technology sometimes and you need to have technical expertise to tune them so they don’t break as often“.

[Perry E. Metzger @perrymetzger]

I have said this before, but to those of us using AI systems to get lots of work done reliably and quickly, the people who post online about how AIs still hallucinate constantly, about how they can’t write code, etc., seem equivalent to people trying to convince you that the car you drive to work every day doesn’t exist.

You tell them things like “but I drive a car. I paid money for it. I buy gasoline for it. I could not possibly be working twenty miles away from home if I didn’t have the car?” and they reply that you are imagining having a car, or that you’re lying because you work for a car company.

It is as though these people live in a completely different reality.

[Konstantin @kauffinger]

Also, non packagist sources can change!

[Konstantin @kauffinger]

Composer/packagist version immutability update is finally live 🚀

[Konstantin @kauffinger]

Dev versions still can change, so if you want to secure against that, or soak time composer plugin lets you guard against that:
https://t.co/MCqCtsWQAE

[Konstantin @kauffinger]

@aschmelyun And scary. I’ve explicitly forbidden any kinds of autofixing stuff from external issues months ago. Text is an executable now :|

[Konstantin @kauffinger]

@peterfox I think it's really handy for learning go as a laravel dev. At least parts of it. Just seeing what the closest representation of Laravel concepts in go are vs php is sick

[Konstantin @kauffinger]

@frankdejonge What I'd do right now to only have 1/8th the code i have at this moment

[Konstantin @kauffinger]

@yacineMTB They used recursive self improvement to create a full stash of gun emojis that they planned to swap with all normal emojis, i heard

[Dimacratia🪿 @codimcom]

@allmeasures

[htmx.org / CEO of Bad Flags (same thing) @htmx_org]

https://t.co/a6kTxZtpf3

[Konstantin @kauffinger]

@FlorianGallwitz Das ganze riecht mir in dem Fall allerdings sehr nach Marketing für die geplanten IPOs. Von RSI zu sprechen wenn die nächste Runde Funding ansteht ist inzwischen Tradition.

[Konstantin @kauffinger]

@Lentils80 > 3k lines of code
> fully functional browser

[Konstantin @kauffinger]

@jrysana @Strife212 We don’t know how to program a simulated brain

[Konstantin @kauffinger]

@tszzl @shakoistsLog Luckily the largest companies on earth are usually right about strategies like this

[Dmitrii Kovanikov @ChShersh]

@ThePrimeagen The last thing you should trust an SWE with is estimations