upvote
Technical people (which is by far the minority of people out there) building personal apps to scratch an itch is one thing.

But based on the hype (100x productivity!), there should be a deluge of high quality mobile apps, Saas offerings, etc. There is a huge profit incentive to create quality software at a low price.

Yet, the majority of new apps and services that I see are all AI ecosystem stuff. Wrappers around LLMs, or tools to use LLMs to create software. But I’m not really seeing the output of this process (net new software).

reply
I worked in an industry for five years and I could feasibly build a competitor product that I think would solve a lot of the problems we had before, and which it would be difficult to pivot the existing ones into. But ultimately, I could have done that before, it just brings the time to build down, and it does nothing for the difficult part which is convincing customers to take a chance on you, sales and marketing, etc. - it takes a certain type of person to go and start a business.
reply
Nobody’s talking about starting businesses. The article is specifically about pypi packages, which don’t require any sales and marketing. And there’s still no noticeable uptick in package creation or updates.
reply
My understanding reading it was that PyPi packages is just being used as a proxy variable
reply
Yes, you are correct. The parent is not following the conversation. They probably didn't even read the article.
reply
There is no money in mobile apps. It came out in the Epic Trial that 90% of App Store revenue comes from in app purchases for pay to win games. Most of the other money companies are making from mobile are front end for services.

If someone did make a mobile app, how would it get up take? Coding has never been the hard part about a successful software product.

reply
Why on earth would you publish and monetize software anybody can reproduce with a $20 subscription and an hour of prompting? Why would you ever publish something you vibe coded to PyPI? Code itself isn’t scarce anymore. If there is not some proprietary, secret data or profound insight behind it, I just don’t think there is a good reason to treat it like something valuable.
reply
> But based on the hype (100x productivity!), there should be a deluge of high quality mobile apps, Saas offerings, etc. There is a huge profit incentive to create quality software at a low price.

1. People aren't creating new apps, but enhancing existing ones

2. Companies are less likely to pay for new offerings when the barrier to entry is lowered due to AI. They'll just vibe code what they need.

reply
I don't think the 2nd point will make a huge impact on software sales. Who is vibe coding? Software developers or business types? They aren't going to vibe code a CRM, or their own bespoke version of Excel, or their own Datadog APM.

Maybe they will vibe code small scripts, but nobody was really paying for software to do that in the first place. Saas-pocalypse is just people vibe investing, not really understanding the value proposition of saas in the first place (no maintenance, no deployments, SLAs, no database backups, etc).

reply
> Wrappers around LLMs, or tools to use LLMs to create software. But I’m not really seeing the output of this process

Because it's better to sell shovels than to pan for gold.

In the current state of LLMs, the average no-experience, non-techy person was never going to make production software with it, let alone actually launch something profitable. Coding was never the hard part in the first place, sales, marketing & growth is.

LLMs are basically just another devtool at this point. In the 90s, IDEs/Rapid App Development was a gold rush. LLMs are today's version of that. Both made developer's life's better, but neither resulted in a huge rush of new, cheap software from the masses.

reply
And SQL was that version in the 80s...
reply
Before LLMs, there were code sweatshops in India, Vietnam, Latin America, etc. and they've been pumping out apps and SaaS products for decades now.
reply
And it was all crap software, no? EDIT: If it was crap, then that is still good for AI.
reply
AI-powered devs are struggling to stand above it so it wasn't all crap, or, AI produced stuff is too
reply
I think this is the great conundrum with AI. I find it's most useful when I build my own tools from models. It's great for solving last-mile-problem types of situations around my workflow. But I'm not interested in trying to productize my custom workflow. And I've yet to encounter an AI feature on an existing app that felt right.

Problem is that all these companies trying to push AI experiences know that giving users unfettered access to their data to build further customization is corporate suicide.

reply
Profit is not everyone's goal.

Me, I'm not just chasing markets; I want to build things that create joy.

reply
Well it’s mostly explained by the fact that most people lack imagination and can’t hold enough concepts about a particular experience to think about how to re-imagine it, to begin with.

Oh and sadly, llm’s are useless for the imaginative part too. Shucks eh.

reply
I share this particular cynicism.

I have a list of ideas a mile long that gets longer every day, and LLMs help me burn through that list significantly faster.

However, the older I get, the more distraught I get that most people I meet "IRL" are simply not sitting on a list of problems they simply lack time to solve. I have... a lot of emotions around this, but it seems to be the norm.

If someone doesn't see or experience problems and intuitively start working out how they would fix them if they only had time, the notion that they could pair program effectively ideas that they didn't previously have with an LLM is absurd.

reply
Also one of those with a mile-long ideas list that I can finally now burn through. I gotta say, it feels good!
reply
Yeah and frankly the innovation would occur irrespective of llm’s.

Would it be harder? Sure. And perhaps the difficulty adds an additional cost of passion being a necessary condition to embark on the innovation. Passion leads to really good stuff.

My personal fear is we get landfill sites of junk software produced. To some extent it should be costly to convert an idea to a concept - the cost being thinking carefully so what you put out there is somewhat legible.

reply
Yes, it'd be better if people kept their inner Oppenheimer in check.

However, I suspect it's much more like the three types of people talking about 3D printers:

- 3D printing jigs and prototypes has completely changed my workflow

- I can't find any more things to print from the vendor provided gallery

- why on earth would I want a 3D printer, you guys are geeks

LLMs are not creating a risk that nihilist socialites will disrupt how device drivers get written.

reply
There really isn't much profit incentive actually, as everyone has access to the same capabilities now. It'd be like trying to sell ice to Eskimos.
reply
Most businesses do not have the capacity to use LLMs to produce software. If you have an idea that you can create into real high quality software that there is a demand for, then you should absolutely do it.
reply
This is probably my favorite gain from AI assisted coding: the bar for "who cares about this app" has dropped to a minimum of 1 to make sense. I recently built an app for grocery shopping that is specific to how and where I shop, would be useless to anyone other than my wife. Took me 20 minutes. This is the next frontier: I have a random manual process I do every week, I'll write an app that does it for me.
reply
More than that. Building a throwaway-transient-single-use web app for a single annoying use kind of makes sense now, sometimes.

I had to create a bunch of GitHub and Linear apps. Without me even asking Codex whipped up a web page and a local server to set them up, collecting the OAuth credentials, and forward them to the actual app.

Took two minutes, I used it to set up the apps in three clicks each, and then just deleted the thing.

Code as transient disposable artifacts.

reply
I posted it recently, but now this works differently https://xkcd.com/1205/

You can get a throw away app in 5 mins, before I wouldn't even bother.

reply
Same energy here. I was sitting on 50+ .env files across various projects with plaintext API keys and it always bothered me but never enough to actually fix it. AI dropped the effort enough that I just had a dedicated agent run at it for a few days — kept making iterations while I was using it day to day until it landed on a pretty solid Touch ID-based setup.

This mix of doing my main work on complex stuff (healthcare) with heavy AI input, and then having 1-2 agents building lighter tools on the side, has been surprisingly effective.

reply
Even if it’s only useful to you it would be super educational to see your prompts and the result.
reply
What exactly were you bale to build in 20 minutes?
reply
Me, and photo editor tool to semi-automate a task of digitizing a few dozen badly scanned old physical photos for a family photo book. Needed something that could auto-straighen and auto-crop the photos with ability to quickly make manual adjustments, Gemini single-shotted me a working app that, after few minutes of back-and-forth as I used it and complained about the process, gained full four-point cropping (arbitrary lines) with snapping to lines detected in image content for minute adjustments.

Before that, it single-shot an app for me where I can copy-paste a table (or a subsection of it) from Excel and print it out perfectly aligned on label sticker paper; it does instantly what used to take me an hour each time, when I had to fight Microsoft Word (mail merge) and my Canon printer's settings to get the text properly aligned on labels, and not cut off because something along the way decided to scale content or add margins or such.

Neither of these tools is immediately usable for others. They're not meant to, and that's fine.

reply
My buddy and I are writing our own CRUD web app to track our gaming. I was looking at a ticketing system to use for us to just track bug fixes and improvements. Nothing I found was simple enough or easy enough to warrant installing it.

I vibe'd a basic ticketing system in just under an hour that does what we need. So not 20 mins, but more like 45-60.

reply
I built a small app to emit a 15 kHz beep (that most adults can't hear) every ten minutes, so I can keep time when I'm getting a massage. It took ten minutes, really, but I guess it's in the spirit of the question.

For 20 minutes of time, I had a simple TTS/STT app that allows me to have a voice conversation with my AI assistant.

reply
That's fine and all, but how much are you ready to pay to Anthropic and OpenAI to be able to do this? Like, is it worth 100 bucks a month for you to have your own shopping app?
reply
It's easily worth the <$1 in tokens from a Chinese model. You don't need frontier reasoning capabilities to make a personalized grocery list app.
reply
That is an excellent question. For me the answer is yes, but I'm unusual.
reply
It's not worth 100 bucks a month for me to have my own shopping app, but maybe it's worth 100 bucks a month to have ready access to a software garden hose that I can use if I want to spew out whatever stupid app comes to my mind this morning.

I'd rather not pay monthly for something (like water) that I'm turning on and off and may not even need for weeks. But paying per-liter is currently more expensive so that's what we currently do.

I think the future is going to be local models running on powerful GPUs that you have on-prem or in your homelab, so you don't need your wallet perpetually tethered to a company just to turn the hose on for a few minutes.

reply
Haha great. I guess my wider point is that most people won't be ready to pay for it, and in the end there will be only two ways to monetize for OpenAI et al: Ads or B2B. And B2B will only work if they invest a lot into sales or if the business owners see real productivity gains one the hype has died one.
reply
I've been getting close to that myself, I've been using VSCode + Claude Code as my "control plane" for a bunch of projects but the current interface is getting unwieldly. I've tried superset + conductor and those have some improvements but are opinionated towards a specific set of workflows.

I do think there would be value in sharing your setup at some point if you get around to it, I think a lot of builders are in the same boat and we're all trying to figure out what the right interface for this is (or at least right for us personally).

reply
I would still be interested even if my personal workflow is different. These things can be very inspirational!
reply
deleted
reply
This sounds chaotic and fun.
reply
Sounds more like satire.
reply
I am easily caught by satire and I have a weakness for buttons.
reply
> I deleted vscode and replaced with a hyper personal dashboard that combines information from everywhere.

Emacs with Hyperbole[0]?

[0]: https://www.gnu.org/software/hyperbole/

reply
You can't mention Hyperbole and not say how you use it. I did not get past the "include the occasional action button in org-mode" phase.
reply
actually the rules say that no one can ever explain what Hyperbole is for
reply
Wdym by "it does something interesting with Claude code I can't do programmatically"?
reply
I'm guessing it's not a hard coded function, the button invokes. Instead it spawns a claude code session with perhaps some oredefined prompts, maybe attaches logs, and let's claude code "go wild". In that sense the button's effect wouldn't be programmatical, it would be nondeterministic.

Not OP, just guessing.

reply
I have had the thought to write little "programs" in text or markdown for things which would just a chore to maintain as a traditional program. (I guess we call them "skills" now?) Think scraping a page which might change its output a bit every so often. It the volume or cadence is low, it may not be worth it to create a real program to do it.
reply
It means he has a girlfriend. And she goes to a different school. In Canada. You've never heard of it.
reply
Perfect analogy actually
reply
... how did that replace vscode?

Do you never open a code editor?

reply
Kind of. I'm finding that my terminal window in VSCode went from being at the bottom 1/3rd of my screen to filling the whole screen a lot of the time, replacing the code editor window. If AI is writing all of your code for you based on your chat session, a lot of editing capabilities aren't needed as much. While I wouldn't want to get rid of it entirely, I'd say an AI-native IDE would deemphasize code editing in favor of higher-level controls.
reply
Well, I’m sharing it. If someone wants an early preview or to work w me on this, the calendly link is on the site:

https://safebots.ai

But it requires A LOT of work to make sure it is actually safe for people and organizations. And no, an .md file saying “PLEASE DONT PWN ME, KTHX” isn’t it at all. “Alignment” is only part of the equation.

If you’re not afraid to dive into rabbitholes, here is how it works: http://community.safebots.ai/t/layer-4-browser-extensions-pe...

reply
Sorry, I'm not sure how this relates to the content of the article. Sounds like an interesting experience, but this is an analysis of the Python ecosystem pre+post ChatGPT.
reply