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

241–250 of 473 posts

Re: Vibe code is legacy code

#241

Post by sysmax »

There's a pretty good sweet spot in between vibe coding and manual coding.

You still think out all the classes, algorithms, complexities in your head, but then instead of writing code by hand, use short prompts like "encapsulate X and Y in a nested class + create a dictionary where key is A+B".

This saves a ton of repetitive manual work, while the results are pretty indistinguishable from doing all the legwork yourself.

I am building a list of examples with exact prompts and token counts here [0]. The list is far from being complete, but gives the overall idea.

[0] https://sysprogs.com/CodeVROOM/documentation/examples/scatte...

Re: Vibe code is legacy code

#242

Post by jongjong »

I love using Claude Code. What happens is if your codebase is bad, Claude will produce bad code which follows the same general approach.

So if the foundation is overengineered, it will produce over-engineered code on top. If your code is super clean and succinct, it will produce code that is equally neat and succinct.

I've used Claude code with my own backend serverless platform and it's like it can read my mind. It's almost exactly how I would do it. It's like working with a clone of myself. I can implement nuanced features and ask for very specific changes and the success rate is extremely high.

On the other hand, I've worked with some other people's code which was slightly over-engineered and it struggled. I managed to achieve what I wanted with a lot of manual intervention and careful prompting but it was a very different experience and the code it produced didn't look like my code. It looked like it was written by a mid-level dev.

I can sense that it's a code completion engine because it clearly doesn't have an opinion about code quality. It doesn't know what metrics to aim for. It doesn't aim for succinctness; just as most human developers don't aim for it either.

My assessment of LLMs is that they don't have a specific intrinsic goal so the goal/motivation is inferred from the text it sees (from the input). If the original author of the code was motivated by creating complexity for the purpose of lock-in factor and career advancement, the LLM will produce code which optimizes for that. Unfortunately, this describes the vast majoy of code which is produced.

You can tell a lot about someone based on their code.

Re: Vibe code is legacy code

#243

Post by rainonmoon »

CharlieDigital wrote: post #67

A story about a non-technical friend: friend vibe coded a SaaS last year and started generating revenue with almost no marketing; all word of mouth and inbound in a niche industry. Used Replit and Supabase to build the thing; I am still really impressed by what he was able to do given how complex the app ended up becoming as he interacted with customers. What I think happened: there are two incumbents in this space t…

That’s not a viable business, it’s a walking liability. Besides which, why would anyone trust your friend (as an investor or customer) ever again when they’ve shown such profound disregard for user data and their IP? If your metric of success is “I have no idea what I’m doing and still made money from it” your friend would have a better time starting a podcast.

Re: Vibe code is legacy code

#244

Post by bschwindHN »

CharlieDigital wrote: post #67

A story about a non-technical friend: friend vibe coded a SaaS last year and started generating revenue with almost no marketing; all word of mouth and inbound in a niche industry. Used Replit and Supabase to build the thing; I am still really impressed by what he was able to do given how complex the app ended up becoming as he interacted with customers. What I think happened: there are two incumbents in this space t…

> Was it worth it? Yes, it is terrible, shoddy, insecure code, but he proved out a viable business with just a few hundred dollars of investment.

Thank god that someone, somewhere, was able to make some money out of irresponsibly releasing software into the world!

Re: Vibe code is legacy code

#245

Post by mschild »

rainonmoon wrote: post #243
CharlieDigital wrote: post #67

A story about a non-technical friend: friend vibe coded a SaaS last year and started generating revenue with almost no marketing; all word of mouth and inbound in a niche industry. Used Replit and Supabase to build the thing; I am still really impressed by what he was able to do given how complex the app ended up becoming as he interacted with customers. What I think happened: there are two incumbents in this space t…

That’s not a viable business, it’s a walking liability. Besides which, why would anyone trust your friend (as an investor or customer) ever again when they’ve shown such profound disregard for user data and their IP? If your metric of success is “I have no idea what I’m doing and still made money from it” your friend would have a better time starting a podcast.

> Why would anyone trust your friend (as an investor or customer) ever again when they’ve shown such profound disregard for user data and their IP?

Plenty of people probably. There are hundreds of businesses that constantly get exposed for massive leaks and/or horrendous business practices yet they're doing just fine. I'd imagine the killing blow in this case would be the stripe key but beyond that they would've likely had no massive issue.

Re: Vibe code is legacy code

#246

Post by efitz »

sshine wrote: post #169
efitz wrote: post #154

If you vibe coded it, why can’t you vibe support it? I think the whole premise of the article is wrong because it compares apples to oranges, assuming that you’re going to maintain “vibe code” in a traditional way.

Because support requires understanding. If you can maintain long-term context for the LLM, then you have an approximation of actionable institutional knowledge. But context is text documentation, and that stuff rots just as much, if not more, when LLMs generate it. Nothing beats having the actual code and the thoughts that went onto it in your wetware. I still remember the hardest problems in my professional life in…

You can have the LLM generate whatever context you need- documentation, developers guide, design patterns, whatever. You can have it do this, at the time you create the app, or later (I have done this for other peoples code I found on GitHub that was commented in a language I can’t speak). You store this by having your LLM coding agent write it down in a file. I do that all the time. It works really well.

Re: Vibe code is legacy code

#247

Post by efitz »

efitz wrote: post #246
sshine wrote: post #169

Earlier quoted context omitted.

Because support requires understanding. If you can maintain long-term context for the LLM, then you have an approximation of actionable institutional knowledge. But context is text documentation, and that stuff rots just as much, if not more, when LLMs generate it. Nothing beats having the actual code and the thoughts that went onto it in your wetware. I still remember the hardest problems in my professional life in…

You can have the LLM generate whatever context you need- documentation, developers guide, design patterns, whatever. You can have it do this, at the time you create the app, or later (I have done this for other peoples code I found on GitHub that was commented in a language I can’t speak). You store this by having your LLM coding agent write it down in a file. I do that all the time. It works really well.

And if you have some kind of special case or something, the LLM will usually comment it when it writes it, but you can also just tell it to do so.

The docs don’t rot because you can throw them away and have the LLM recreate them at any time. And they’re not for you, they’re for the agent.

It’s a completely different mindset.

Re: Vibe code is legacy code

#248

Post by zerkten »

JackFr wrote: post #235
eddythompson80 wrote: post #120

Earlier quoted context omitted.

Who else would spend the time and effort to figure out you leaked your stipe key to your front end? Sure people have bots to do that, but it’s kinda unbelievable someone would run such a bot on their vibe coded website.

I have a strictly hobby web app that I work on. 6-7 years ago I inadvertently pushed AWS email service credentials to GitHub. Half an hour after the push I got an email and text from GitHub that I had exposed credentials. I quickly logged in to my AWS to turn off the service, to see that AWS had suspended that service because the bounce rate on the 80000 emails sent in that 15 minute period was too high. It was crazy…

>> It was crazy just how fast it was exploited.

People underestimate the speed, but also the number of pivots that advanced attackers will make. Sure, these kinds of problems are easy to exploit, but with major organizations that employ reasonable defenses, the attackers will pivot through 50+ exploits/machines/layers to get to the target(s). This can take place over weeks or months.

Re: Vibe code is legacy code

#250

Post by lifeformed »

extr wrote: post #35

IMO, the time of "code as math" is over. No sufficiently large software system that interacts with the real world is provable to be correct like a mathematical statement is. They are all complicated, engineered systems that are backed by a mix of formal guarantees, earned design principals, experimental testing, rules of thumb, acceptable performance envelopes, etc This is what all software will become, down to the s…

This assumes software is a thing you build once and seal it off when it's finished.

What happens when you need to modify large portions of it? Fix security issues? Scale it up 20x? You can throw more tokens at it and grow it into a monstrous hulk. What if performance degrades due to its sheer weight?

I know humans aren't perfect and are capable of writing really bad unmaintainable code too, but this is just embracing that more. This feels like going further down the same route of how we ended up with 10MB websites that take many seconds to load. But yeah it will probably win over the market.

Post reply on HN