[with]

the real takeaway is buried at the bottom: "the magic isn't in the input, it's in the system around it." random keystrokes producing playable games means the input barely matters anymore. we're basically at the point where the engineering is in the scaffolding, not the prompting.

[avaer]

> we're basically at the point where the engineering is in the scaffolding, not the prompting.

This still required prompting, and not from the dog. Engineering is still the holistic practice of engineering.

[cardanome]

That also shows the delusion of some people that believe their vibe coded projects have any value.

If generative AI improves at the rate that is promised then all your "promting skills" or whatever you believe you had will be obsolete. You might think you will be an "AI engineer" or whatever and that it is other people that will lose their job, that you are safe because you have the magic skills to use the new tech. You believe the tech overlords will reward you for your faith.

Nope. You are just training your replacement.

No one will buy your game that you vibe coded. If the tech were good enough to create games that are actually fun then they would just generate their own games. Oh your skill? Yeah, a dog can do it.

Yes people will cope by saying but oh the whole initial prompt and setting it all up was still hard but yeah currently. The tech will improve and it will get more accessible. So enjoy the few months you are still relevant.

Of course there is reason to believe that you can't scale up LLMs endlessly and bigger models hit diminishing returns. In fact we might already be seeing this. So there is an upside but then again when the AI bubble pops and the economy crashes you will be out of a job all the same.

[mikkupikku]

Sounds great to me. Software devs might lose their jobs but billions of people will be empowered to spin up whatever software they need on demand. This is the future I dreamed of when I was a kid, and I'm not so cynical as to let the dying of a trade sour me to this objectively incredible technology.

[DrewADesign]

Nobody gives a damn about the dying of a trade. People don’t want their house foreclosed on when they lose their income, or their cancer to kill them when they lose their health insurance, to move an elderly parent into a cheap shitty old folks home because they can’t afford home health care, or not be able to pay for their kid to go on that school field trip.

This would all be pretty fucking swell if the fundamental problems this could cause were even considered before hitting the gas. Instead, you’re going to have a shitload of people with ruined lives, but as a consolation prize, they can vibe code stuff! Wowee!

[woah]

This very forum was founded by a VC who had great success recruiting 22 year olds with fancy diplomas to automate away the job of the guy who copied the numbers from the TPS report pdf attachment into excel.

I didn't see people on here ranting and taking up the flag of revolution for the TPS report excel paster guy's job that they were automating away with their web2 SaaS startup.

But wait- that guy himself was automating away the job of the lady who used to physically Xerox the TPS report and put it in the filing cabinet down the hall, but that lady was automating the job of the secretary who used to re-type all those TPS reports.

It's automatic filing cabinets all the way down, and ranting because your little slice of the filing cabinet automation machine has been made redundant is a bit silly.

[mikkupikku]

You act as if this is the first time in history technology has wiped out a trade and made people scramble to sort out their lives. No, this has been happening over and over again throughout history and at a rapidly accelerating pace since the industrial revolution. Why should we ask it to slow down on behalf of programmers, when it never did for anybody else? Don't pretend you didn't know this was a possibility when you got into tech in the first place. You might have to downsize your life but humanity as a whole will be better off.

[cardanome]

The industrial revolution resulted in children being worked to literal death. Of people toiling 16 hours ad day and living in cramped up spaces without any windows and barely any hygiene. It brought suffering on a scale never seen before.

Organized labor movements managed to fight back and improve conditions somewhat but will we be able to do it this time?

Humanity will not profit from generative AI, tech billionaires will. It is based on the theft of human labor of millions of programmer, artists and writers without any compensation. If left unchecked it will destroy the environment, any form of democracy, our mental health. It will cause mass unemployment at a grand scale.

Could it be in theory used for good? Maybe. As the current political situation stands it will cause massive suffering for the majority of people.

[paulryanrogers]

> You might have to downsize your life but humanity as a whole will be better off

This assumes that there will be other jobs to get. If AI replaces a large enough segment of office jobs then huge portions of the population will be unable to afford essentials like food and healthcare.

[code_for_monkey]

it also assumes humanity will be better off because people can vibe code stuff, which is a massive stretch already.

[mikkupikku]

Walk into a staffing agency, ask for a job. They'll give you a list, pick the one that sounds the least disagreeable. Show up on time, every day, for at least two or three months and you'll convert the temp position into a full time job.

It's literally that easy, showing up reliably is a super power that puts you in the 90th percentile of workers these days. The job probably won't be as comfortable as sitting on a comfortable chair in an air-conditioning office wiggling your fingers at a computer, but so what? Other people make it work, so can you. Man up.

[DrewADesign]

Your glib dismissal of the real effects of those technological upheavals shows you haven’t actually looked into this. You should probably tamp down that smugness until you find out.

[mikkupikku]

Did you think about it before you got into the tech industry? You should have, technology has been wiping out jobs since forever but you got into tech anyway. Live by the sword, die by the sword. Except you needn't actually die, just walk into a staffing agency and ask for a new job. I have done so before, and will do so again. I have prepared for what is to come, saw it coming 20 years ago and saw the imminence of it when GPT-2 was released. I have sympathy for other kinds of white collar professionals who never could have anticipated these kind of developments, but technologists? Give me a break. You knew, or should have known, that technological developments in this domain were likely.

[lelanthran]

> This is the future I dreamed of when I was a kid, and I'm not so cynical as to let the dying of a trade sour me to this objectively incredible technology.

I feel that you should take a longer-term view of things...

If an AI can vibe code from the requirements of the average white-collar worker, we're not talking about the death of a trade. Or even two trades. We're talking about the death of almost all white-collar jobs.

Development paid a lot more than other white-collar work because it was harder, and fewer people could actually do it. How fast do you think the easier work will get replaced if the hardest one is replaced? For the remaining white-collar roles that consist solely of skills achievable by a border collie, how much do you think they'd pay?

[imtringued]

Yeah this is one of the most perplexing things about this "dying of a trade" narrative.

Software development isn't just the act of producing a deliverable that is being gate kept by people who use their own body. Software development has become specialized enough that it is often highly domain specific. To replace the "trade" you need to automate the software part and the domain knowledge part. If you can do both, you've automated every single white collar job in existence.

Since it is possible to write software for machine learning, which is used to solve problems that classical algorithms failed to solve, the amount of problems that cannot be solved using software is shrinking rapidly. If you can write software for any domain, you can solve any domain by using said software.

General purpose software generation can be reduced to AGI completeness. In a way, it is the last job that can be automated.

[malicka]

> Billions of people will be empowered to spin up whatever software they need on demand.

… so long as they have the money, and the power grid survives the overtaxation.

After all, why bother encouraging a culture where people are genuinely empowered to tweak and create their tools? Why encourage a culture of exploration, of playful cleverness? What use is there to being a hacker, of sharing knowledge?

It’s definitely much easier, more sustainable, and more fulfilling to have server farms adjacent nuclear reactors make your calculator app for you.

[mikkupikku]

Hardware will get better, local models will get better, it won't be very long until running a coding agent off a few solar panels on your roof is viable. And if this isn't true, then the future I envision probably isn't going to happen and you lot can stop worrying about your jobs.

[nehal3m]

The way I see it is either it will work (which I doubt, but whatever) and I'm out of a job, or it doesn't work and the economy comes crashing down harder than ever seen in history and I'm out of a job.

[mikkupikku]

The bubble popping is a guarantee even if AI does end up working. The internet certainly works but that didn't stop the dotcom bubble from crashing. Live frugally below your means to save up as much extra money as you can, and do whatever else you think prudent to keep your ass covered.

[malicka]

I’m not worried in the slightest about jobs. A world where acceptable slop is created on demand is one where people lose their curiosity and drive. Why bother practicing drawing if you can prompt an acceptable facsimile of your image to life? Why bother learning to program if you can prompt an acceptable slop program onto your machine?

Killing the drive to learn and explore is not empowering; it is fundamentally disempowering.

[Thanemate]

>billions of people will be empowered to spin up whatever software they need on demand

As long as: 1. They have access to a computer 2. They have affordable access to a capable language model 3. Someone will actually care about using their output instead of simply spinning up their own custom version of whatever idea they have

The number 3 is something many people miss, especially on HN: Why would I want to use YOUR software if it's easy for me to cook up my own? Perhaps out of efficiency or lack of time, in the same way I order pizza instead of baking my own when I'm tired or can't be bothered to bake pizza.

Then the software becomes truly throwaway, in the same way takeaway is, and everything is a greenfield project because rewrites are literally easier and faster to make than patching up existing stuff.

[mikkupikku]

> Why would I want to use YOUR software if it's easy for me to cook up my own?

You're still in the mindset of thinking about software as something you sell to other people. Forget that crap. Software will be something you summon on demand to solve a specific problem you have. As long as people have problems that computers are good at solving, they'll keep using computers. What likely won't continue is computer programming as a career, but so what?

[kevinsync]

Everybody keeps parroting stuff along these lines but never take into consideration:

1. the stark, obvious reality is that most people don't know how to actually use computers! They know what 10 steps they need to take on a computer, in a specific sequence, to complete their task, but anything beyond that is too much.. and they need taught those 10 steps (as well as have it documented somewhere) for it to ever stick

2. not only not know how to use, but simply don't use computers at all! They've got phones and tablets and smart TVs and talk to their Bluetooth speakers and shit but they aren't sitting down at a desk with a keyboard and mouse and using a computer. I'd wager that of the percentage of people who do, an overwhelming majority is doing this primarily at their job to complete work tasks

3. companies with more than 10 employees are absolutely not going to be running to Claude to spin up custom programs to do their work. It's just not happening. Not to mention you can rarely even install unapproved, AAA-quality software on company-issued MDM'd hardware, let alone something generated out of thin air that has a ton of dependencies, no installer, no packaging, not code-signed, etc

4. that pizza you're ordering? You'd never order again if it was a roll of the dice with regards to what you receive. When you pay for two extra large thin crust pies with everything and are delivered some cheesy bread, a 2-liter of Coke and some brownies, your wig will completely and fully split and you'll never patronize that establishment again. Claude absolutely can make you exactly what you order, if you know what to order, and why, but most people don't

5. consistency and determinism matter to businesses, and to people -- both home users and professionals. Most people get stymied by the simplest tasks on a computer, tasks that have deep, instantly-available answers available with a single Google search or ChatGPT session. Guess what they do instead? Give up, and then ask IT or "a tech friend" for help ... how am I going to help you troubleshoot software I've never seen before? That NOBODY has ever seen before? That I can't even install because it only exists as a dev build in a single folder on your hard drive? How are you going to take that program with you when you upgrade your laptop? What if they didn't use git and their computer dies? Ask Claude to remake it? Will it be the same? Do they even know what git is? Do they even know where the folder on their computer that holds the files is located? Or what, you had Claude build a hosted product? Where's it hosted? How much does it cost every month? What if it gets hacked? I could go until my head explodes with all the hypotheticals

6. professionals pay for convenience and predictability, as well as to offload risk and unnecessary labor onto third parties. This will never change. Companies have been worrying about and hedging against "the bus problem" for decades, and vibe-coded software creates the ultimate bus problem: not only are you the only one likely to be in possession of the program in question, you're the only one who has ever seen it, know how to use it (which is different than knowing how it works), and it dies with you. Fine for a personal gadget, but a non-starter for a tool that a business or professional relies on to make their real money

I could go on and on, but you probably get the point. Takeaway food is both throwaway in a different sense than vibe-coded software, and infinitely more accessible to the average human. People are still going to pay for SaaS, still going to buy software, and still going to build software. In fact, I'm starting to think we'll see less open-source contributions and more closed-source, for-profit software released than ever before as a result of Claude and Codex, rather than a complete flattening and decimation of this industry. I think people in software will try to become more entrepreneurial as a result of corporate job loss. I also think that a byproduct of this coming tsunami of new commercial products is that the overwhelming majority will be low quality noise, and the proportionality of signal -> noise will remain largely unchanged. I'd use social media as an example (a staggering amount of people show up and try to break through, a very small percentage actually do) but IMO you see it in any industry: there can only be a few outsized successes in anything at any given moment in time (but also a not-insignificant amount of medium-sized success that flies under the mainstream radar)

I dunno. Maybe I'm full of shit, but I still think it's absolutely bonkers to think that the software industry is over because every person will just become sovereign groundskeepers of all of their own bespoke software. We can't all be our own bank, lawyer, doctor, mechanic, fitness trainer, software developer, chef and bodyguard, while also dealing with the other stuff that are our primary responsibilities! And that means that as long as society doesn't fully collapse into widespread economic ruin, and we aren't all unemployed, desperate, violent marauders trying to survive in District 9, there will be plenty of opportunities out there in the software space. They might just look a little different than they used to, and you'll have to go out and get them

[jarhag]

Shills like yourself indeed need not worry. You just find a new scam to shill.

[code_for_monkey]

sure, a bunch of people will lose jobs, but at that trade off everyones dog can vibe code Royal Frog, a 4 level unwinnable game where play as king frog, eating peasant flies.

[ajspig1]

+ Also the fact that the Memory.md file was a hindrance to the quality of output

[cezart]

Depends on the desired output. The author wanted variability, for which Memory.md was an obstacle. Another project might need consistency.

[cleak]

In this case yes, but the real takeaway is to pay attention to Memory.md. If I had a particular game in mind and it latched on early to a style I didn't like, there's no guarantee it would update the memory as I try to change the style.

[otabdeveloper4]

> the engineering is in the scaffolding, not the prompting

Well, yes. Feeding random tokens as prompts until something good comes out is a valid strategy.

[yoyohello13]

Simulated annealing for game design

[cardanome]

Not that I condone any form of gambling but I would rather play actual slot machines instead of spending hundreds of dollars on tokens in hopes that the AI blesses me with anything useful.

[otabdeveloper4]

This sort of thing is usually done on the frontier of LLM research. (Role play with local models.)

[wazHFsRy]

I think this misses something. The output here is something not the thing. Yes the scaffolding is important, but the requirements are even more important. You need crystal clear requirements + great scaffolding and then the implementation becomes mechanical.

[krupan]

This is amazing because it's the same logic and argument about how to do good software engineering that's been around for 40 years. If you just write good enough requirements, a good enough, detailed specification, then your software team can't fail, even if they are low-cost engineers from a developing nation. It's the classic Waterfall method.

That was totally upended by agile, that emphasized that yes, a clear, unambiguous specification is needed, and the best language for that is a programming language. Don't waste time writing a detailed spec in English, get right to writing it in code that you can execute and get immediate feedback on.

Now people want LLMs to write the code for them, so they are back to saying we just need to give the LLMs clear enough direction, a clear specification. It's amazing to witness history not exactly repeat itself, but very clearly rhyming

[azeirah]

Sorry but what do you mean crystal clear requirements?

I don't particularly think "y7u8888888ftrg34BC" would pass as a crystal clear requirement at my workplace :<

Do you mean something different?

[dormento]

> y7u8888888ftrg34BC

This is more information than the average users gives you when requesting new features.

[wazHFsRy]

I mean you get a random game in the authors example :) But in real life you do not want a random game. That's what I mean, you need the great scaffolding + exact requirements. Then the prompt to do the implementation does not matter too much.

[withinboredom]

Recent research has shown that specs actually degrade output, so I’m more inclined to think you’ve got the slot machine mentality mentioned up-thread.

[dovirk]

Source?

Last I checked, specs improved output quality. There are hiccups with complex, difficult specs but I guess that much is obvious.

[withinboredom]

https://arxiv.org/abs/2602.11988

[testaccount28]

this would be a more insightful comment if the output wasn't itch io shovelware.

[randomtoast]

> the magic isn't in the input, it's in the system around it.

It isn't [this], it's [that]. Is AI slop, just saying.