Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Vibe code is legacy code

blog.val.town

Post reply on HN
Print view View original

441–450 of 473 posts

Re: Vibe code is legacy code

#441

Post by api »

__MatrixMan__ wrote: post #391
api wrote: post #197

Earlier quoted context omitted.

In that world we’d just be transitioning to 32-bit software and still running MS-DOS since it’s certified. Linux would never ever have broken through. Who can trust code developed by open source cowboys? Have we verified all their credentials? There are some industries where the massive cost of this type of lock down — probably innovation at 1/10th the speed at 100X the cost — is needed. Medicine comes to mind. It’s…

I disagree. Mature open source projects last long enough without significant disruption to still be relevant after they make it onto the certification exam. Products, not so much. Investing time building familiarity with proprietary software is already a dubious move for a lot of other reasons, but this would be just one more: why would I build curriculum around something that I'm just going to have to change next ye…

You're assuming the process would not be instantly subjected to regulatory capture by for-profit companies and by universities with an interest in inserting themselves into the required licensure pipeline.

Microsoft in the 1990s would have used the regulatory and licensure process to shut down open source. They tried to do it with bullshit lawsuits.

Re: Vibe code is legacy code

#442

Post by oxag3n »

Legacy code may be very different - I saw legacy libraries implemented in C, with macros APIs that just worked.

Same applies to "live" code with owners still within a team - it can be extremely bad.

The issue with LLMs is you don't know what you get. It can produce a very robust routine, that will just work and have a clear API, or it will look like it, but will have a terrible design decision that will lead to an API redesign.

Re: Vibe code is legacy code

#443

Post by SecretDreams »

exe34 wrote: post #439
SecretDreams wrote: post #409

Earlier quoted context omitted.

> Being in the possession of a password or key implies having permission to use that key So if I get your house key I can use your bathroom? Seriously, what hill are you trying to die on here?

Depends - did I hand it out at the street corner?

Nah, you left it under your door mat.

Re: Vibe code is legacy code

#444

Post by throawaywpg »

sysmax wrote: post #438
danielbln wrote: post #314

Earlier quoted context omitted.

I'm still finding the right sweet spot personally. I would love to only think in architecture, features and interfaces and algorithms, and leave the class and function design fully to the LLM. At this point this almost works, but requires handholding and some retroactive cleanup. I still do it because it's necessary, but I grow increasingly tired of having to think to close to the code level, as I see it more and mor…

Most of the AI hiccups come from the sequential nature of generating responses. It gets to a spot where adhering to code structure means token X, and fulfilling some common sense requirement means token Y, so it picks X and the rest of the reply is screwed. You can get way better results with incremental refinement. Refine brief prompt into detailed description. Refine description into requirements. Refine requiremen…

hello, im interested in the GUI options you mentioned!

Re: Vibe code is legacy code

#445

Post by genewitch »

thesuperbigfrog wrote: post #151
genewitch wrote: post #82

Earlier quoted context omitted.

We're seeking angel investors for our startup that does this, we train models with "assembly" that does specific things and also complete "programs"; the end goal to prompt and it outputs executables. It's farther along than quantum computing at solving real problems, for instance, it can factor "15". this is like the third time i've mentioned this on HN (over a year.) Apparently everyone else is too busy complaining…

>> It's farther along than quantum computing at solving real problems, for instance, it can factor "15". Dogs can also factor 15: https://eprint.iacr.org/2025/1237.pdf

> We argue that the amount of sleight-of-hand used in this method is no more than typical quantum factorisation sleight-of-hand.

great!

Re: Vibe code is legacy code

#446

Post by eddd-ddde »

fc417fc802 wrote: post #345
AlecSchueler wrote: post #304

Earlier quoted context omitted.

I don't actually see the difference. If someone writes the code and understands it but then becomes unavailable, what's the practical difference with no one having understood it?

Someone at some point had a working mental model of the code and a reputation to protect and decided that it was good enough to merge. Someone vetted and hired that person. There's a level of familiarity and history that leads others to extend trust.

The way I see it is that LLMs and humans are not inherently different. They are simply on different segments of a very complex spectrum of sensory input and physical output. Over time this position on the spectrum changes, for both LLMs and humans too.

With this in mind, it's all matter of what are your metrics for "trust". If you are placing trust on a human employee because it was hired, does this mean the trust comes from the hiring process? What if the LLM passed went through that too?

About familiarity and history: we are at the point were many people will start working at a new place were the strangers are the humans, you will actually be more familiar and history with LLM tools than actual humans, so how do you take that into consideration?

Obviously this is a massive simplification and reduction of the problem, but I'm still not convinced humans get a green checkmark of quality and trust just because they are humans and were hired by a company.

Re: Vibe code is legacy code

#450

Post by escapedmoose »

0x500x79 wrote: post #141
pyman wrote: post #2

Something interesting is happening. A false narrative is spreading online, pushed by people who know little about engineering, and others who should know better. They claim junior devs are now 10x more productive, and project managers are shipping code themselves. Now, close your eyes for five seconds and try to picture what that code looks like. It's 100% legacy, disposable code. The problem isn't AI, or PMs turning…

I had a PM at my company (with an engineering background) post AI generated slop in a ticket this week. It was very frustrating. We asked them: "Where is xyz code". It didn't exist, it was a hallucination. We asked them: "Did you validated abc use cases?" no they did not. So we had a PM push a narrative to executives that this feature was simple, that he could do it with AI generated code: and it didn't solve 5% of t…

I’ve recently had a couple people try to help me fix code issues by handing me the results of their AI prompting. 100% slop; it made absolutely no sense in the context of the problem.

I figured the issue out the old-fashioned way, but it was a little annoying that I had to waste extra time deciphering the hallucinations, and then explaining why they were hallucinations.

Post reply on HN