Waste… I can’t stop thinking about the waste of human talent and potential. The waste of resources to run AI data centers. The waste of the now old school CS ethos. Yea, wasteland checks out.
I really want to host a vibe coding competition and see what can actually be made with these systems. Like if we’re doing insane token spends, it better be in service of creating amazing stuff. Can we make an entirely new programming language? Can we make an OS?
That would be awesome to see! If there was some prize pool like $5,000 to build an operating system through vibe coding tools only and then people could stream themselves on twitch and you could have "sorts desk" type commentators who are collating all that together. I'd watch that and donate a hundred bucks.
That’d also be an interesting data point! What’s the upper limits of vibe coding? Can you vibe code rust? What about an entire programming language toolchain? How about an ecosystem? Can you make a parallel npm?
In all honesty, if you scoped this well, one of the big players in the LLM space could definitely host a big marketing event on this spin. Get together a bunch of well known industry folks, have them vibe code a working <thing> in a given time constraint, presentations and prizes, lots of marketing.
I am designing one, aimed at Claude Code and other AI Coding Agents, and getting the first version lex/parser/compiler was an afternoon project. It was initially a TypeScript toolchain generating TypeScript code.
I keep adding things here and there, a couple hours everyday. Then after about a week I decided to switch the toolchain from TypeScript to Rust, how much work? A 5 minute planning session and a ~20 minutes implementation phase.
Possible, yes. Easier? I tried to search for YouTube videos of people doing amazing things at blazing speed using Gas Town about a month ago, and couldn't find any. I for sure didn't want to spend hours reading and learning something that I don't know if it even works?
Does anyone have like, projects built using it? I couldn't find "look at the output" types of videos or articles or repos, only "look at the input" types of posts about it.
The guy to watch here is https://github.com/Dicklesworthstone . He's rewritten SQLite in Rust with fixes, written his own Rust async engine with fixes that Tokio doesn't have, generated an insane number of tools for agentic orchestration (indexing of all sessions across all harnesses, on-demand skill storage, agent mail), and is currently building out agent orchestration terminal multiplexer stuff.
Source: been watching both these guys closely, as I've been building my own agent factory focused on security + learning: https://github.com/mieubrisse/agenc
It's interesting progression from Gas Town, but it seems like the bottleneck is still translating ideas into actionable input/output frameworks for various agentic tasks.
Also there's the issue of how to identify systems-interface problems and posting those tasks for completion as well. No guarantee that a totally federated system will not solve interfacial issues faster than they generate them without feedback and oversight.
Reads like a scam. Obfuscatory language, outsized claims on future impact, excited opportunity advertisement, first-mover advantage, "no time for the rulebook, it's an inch thick!".
Hilarious too that he uses the word "wasteland" for something that's supposedly good. Perhaps it is a double coding where all of us normies mock it but the FOMO-blinded have their own private reading and say the rest of us are totally wrong.
Reminds me of those 419 emails where the grammar is bad, the story makes no sense, but hey there really are people who expect $10 million to fall out of the sky because they already had $10 million fall out of the sky on them.
I find it amusing that the crypto crowd has yet to recognize their echo chamber and decided to bring it with them during their big pivot to ai. Their culture is offputting
People should stop giving Steve Yegge as much attention as they do.
It's slop on top of slop on top of slop. It's not even quality slop. Apart from bloviated self-aggrandising blog posts, the ideas are trivial, and the execution is beyond horrendous.
Look beyond the ChatGPT-generated terminology to see the supervisors, loops and worker processes of Gas Town. Now it's a freelance board with AI agents as freelancers. But sure, polecats, mayors, stamps and character sheets. Whatever floats your upcoming crypto rug pull [1].
This is what he writes: "build stuff really, really fast. So fast that your biggest problem will be ideas." We've yet to see ideas built using the slopcoded monstrosities.
AI got to him awhile ago, I'm afraid. Been telling stories about these gastowns, zero code review and thousands of lines of code and thousands of dollars in burned tokens since last year.
What's the incentive for anyone to participate in this? It seems wildly expensive (in tokens), high mental overhead to understand, high hardware cost. So what's the upside?
Is it for people with too much money and time on their hands to flex on Github?
I could at least understand the pitch if there was a crypto scam attached, or if folks were getting paid somehow (which might be an interesting social/AI experiment). But that doesn't seem to be the case.
dolt is cool until you hit merge conflicts, need non-simple migrations, or you have a process that ETLs data before putting it in dolt in bulk (huge changesets because one column changes in every row)
separate company, not aware of any association, dolt has been around for a while
This is a classic example of a 'Solution in Search of a Metaphor.'
Strip away the 5,000 words of Fury Road fan fiction and you’re left with a multi-agent wrapper for Claude Code that effectively automates the generation of technical debt. It feels like Yegge is trying to brand 'shoveling tokens into a furnace' as a new paradigm, but the cognitive overhead of learning his proprietary 'lore' just to manage a tmux session of LLMs is a massive net loss in productivity.
We don't need a Wasteland; we need tools that actually improve the signal-to-noise ratio, not industrial-scale noise generators.
I think this is part of the point, perhaps? I get strong Urbit vibes from the Gastown fanfic. Take a relatively simple concept, but invent an entirely new vocabulary to describe it. It satisfies the creator's ego, and acts as a filter for non-believers. People need to buy in to the creator's vision and commit to learning the lingo in order to engage. People who don't get dismissed as lacking the capacity to understand. In other words, cult behavior.
Every Yegge post about AI reads like a Music Man style con job, but he’s got Silicon Valley startup founders salivating and pushing his book to their employees.
With LLM-based tools that inherently rely so much on the semantics of language, I wonder if there will be differences in code generated for the "wanted board" in the "Wasteland", compared to the "task list" in the "public square" or the "wish list" in the "Utopia".
> The Wasteland is a way to link thousands of Gas Towns together (...) to build stuff really, really fast. So fast that your biggest problem will be ideas.
This reads like a speech from Pete Hegseth.
"Let's do war! Hard! Let's build stuff! Stat!"
Build what? Fight for what?
"The hell if we know! Just get busy dropping bombs, or "stamps" or whatever! Faster!"
In the end, all that's left is, indeed, a wasteland.
Yesterday I wanted to change a white background to transparent on some clip art. I’m still learning Affinity so asked Google Gemini Nano Banana PRO 2. The output looked ok at first but the grey squares were a little off. They didn’t make a perfect grid. I opened it in mspaint and was able to erase the grey squares. It didn’t change the white background to transparent, it just drew an array of grey squares, but only good enough for a first glance. I have no idea how these AI tools can make anything of use if left to their own devices.
"We pay for 10000 AI bots and it makes awesome software, which is invisible and undetectable by anybody except me. It gives me a sense of pride and accomplishment."
It's saddening to see folks here on HN direct so much contempt towards this. I have no intention of messing with Gas Town or Wasteland, but I think it's cool that folks like Steve with lots of money and time are using it to build stuff they find cool and interesting. I doubt these projects are the future of AI agent orchestration, but I do think they're probably going to help us collectively learn better how to work with AI.
And if they don't, so what? Who cares? Why be angry at Steve for playing around with a fun hobby project that goes nowhere?
Regardless of whether this particular project goes anywhere, it's at least very interesting that Yegge has discovered a way to make multi-agent setups work better. Giving them discrete personas ("you are a senior database engineer with 30 years of experience") and narrower scopes makes them much more effective. This was surprising to me but makes a lot of sense in retrospect.
The part that always struck me as weird about this stuff is that all of these "agents" with their "personas" are the same baseline LLMs with the same training ultimately, just told to basically pretend they're different. How far can that really get you?
I'm not actually a database engineer with 30 years of experience. If somebody demanded that I pretend to be one, I guess I'd give it a shot, but I would expect any actual employer would be able to tell that I don't have the level of knowledge and experience that you'd expect from somebody like that.
If the base LLM actually has the knowledge of all of these specialties, why can't it just apply them all at once, instead of needing to be told to I guess pretend to be only one of them.
I see a lot of folks lamenting how Yegge has ignited a fire under leadership for egoless, factory worker style engineers. I don't think many would care much otherwise, but his post here is lamenting AI is not enough of a factory worker yet. On a site mostly populated by folks who put most of their lives into becoming skilled professionals, hearing "we think your work should be the kind of work we send to the cheapest, least-developed places on earth" like factory work is disheartening. Of course, it mostly seems like Yegge is here to make money off meme coins and selling Dolt plus his vibe-authored book, so why would anyone take this seriously?
I mean, he can't explain what he's building except pickaxes to make more pickaxes, so it's a bit suspicious. It's just incredible how much impact he has had with this little hustle, given his products are basically turtles all the way down.
> Why be angry at Steve for playing around with a fun hobby project that goes nowhere?
People dislike scam artists, hype artists and bullshitters. Especially when said artists had something actually useful to contribute once upon a time. E.g. Yegge's Platform Rant [1] is still required reading IMO.
Now he's uncritically and unapologetically pushes extremely low quality level AI slop while first trying to prop up Amp, then trying to sell a book, then trying to sell a crypto scam, now trying to sell a vibe-coded database. All the while proclaiming his projects have the basest of basic ideas but somehow need hundreds of thousands of lines of AI-generated low quality slop code to barely function.
The contempt here is the same as for idiots who uncritically run clawdbot and other AI bullshitters and grifters.
Compare this to @simonw who constantly evaluates coding agents, explains what he does in a coherent clear language that doesn't use ChatGPT to invent new inane terms for existing things, and stands behind his work and motivations: https://simonwillison.net
You can be part of the new reputation economy, IF you can afford Multiple Claude Max subscriptions and use all your tokens for it.
Kind of like "getting good" at a collectible card game. It's more weighted toward whether you have money (and are willing to spend it) than anything else.
The best software is both the right software, and high quality. It’s bloody hard to do!
So, we do requirements gathering, and try to get the details right (TDD, continuous delivery etc).. it’s surprising to see Gas Town do none of this and optimistically hope agents will converge on good software by just throwing tokens at a wall and hoping it sticks.
So anyway, that’s what they’re ignoring. What are they actually doing???
It’s all for themselves.
Gas Town’s “product” is the warm fuzzies it gives people with money to burn, warm fuzzy feelings of being “at the frontier”. It’s a luxury product for nerds, and the only ones making money or selling anything are the big labs. There’s zero output or benefit to society because that’s simply not the point.
I think Gastown is truly special, but I wanted something more focused on learning as I think that's the real bottleneck. So I built AgenC to make it trivial to roll learnings back into your Claude.
I mostly use LLMs in a zero-touch way - I never actually edit code and I almost never read it. But I do still dive into the details by exploring it with targeted questions through the LLM. Sometimes I go through ridiculously long sessions to get the LLM to "see" the correct/optimal/simplest/etc solution itself. There are many times when it simply never gets there no matter how close I get the horse to the water. I recently did one of these sessions yesterday and it reinforced my impression that systems like gastown and pure Ralph loop style is just not ever going to have the quality I'm looking for, and it's going to cost a lot of money not to get there.
I've honed a relatively decent flow that requires interaction from me for important parts (mostly) while making its own decisions at the not-important parts (mostly). This results in being able to send the agent off on an hours-long dev cycle and have relatively decent results after that need a few minor fixes. I think this is the best style for the current generation of AI
throwup238 | a day ago
This is all just performance art at this point, right?
jackyinger | a day ago
zingababba | a day ago
RGamma | a day ago
xnorswap | a day ago
I've not seen anyone take Gas Town seriously or even semi-seriously in the way people take Openclaw semi-seriously.
Gas Town certainly seems like a concept for performance art. The original blog post read like it, and so does this.
andrew_lettuce | a day ago
righthand | a day ago
benwad | a day ago
conartist6 | a day ago
dgathercole | a day ago
qsort | a day ago
hardwaregeek | a day ago
rustyboy | a day ago
steveklabnik | a day ago
I have seen both of these already. I've done the former personally, and I've seen links to at least kernels for the latter.
(I didn't do it via gastown, just regular old "use Claude".)
hardwaregeek | 23 hours ago
karmakaze | a day ago
Sevii | a day ago
avaer | a day ago
Etheryte | a day ago
ramesh31 | a day ago
This is trivial in a few hours with Claude Code
inerte | a day ago
I keep adding things here and there, a couple hours everyday. Then after about a week I decided to switch the toolchain from TypeScript to Rust, how much work? A 5 minute planning session and a ~20 minutes implementation phase.
Trivial stuff indeed.
xnx | a day ago
inerte | a day ago
Does anyone have like, projects built using it? I couldn't find "look at the output" types of videos or articles or repos, only "look at the input" types of posts about it.
g-b-r | 19 hours ago
The question is the degree to which they can produce original things.
mieubrisse | 18 hours ago
Source: been watching both these guys closely, as I've been building my own agent factory focused on security + learning: https://github.com/mieubrisse/agenc
condensedcrab | a day ago
Also there's the issue of how to identify systems-interface problems and posting those tasks for completion as well. No guarantee that a totally federated system will not solve interfacial issues faster than they generate them without feedback and oversight.
andrew_lettuce | a day ago
plagiarist | a day ago
robotmaxtron | a day ago
RGamma | a day ago
samothrace | a day ago
I'm good, thanks.
PaulHoule | a day ago
Reminds me of those 419 emails where the grammar is bad, the story makes no sense, but hey there really are people who expect $10 million to fall out of the sky because they already had $10 million fall out of the sky on them.
verdverm | a day ago
troupo | a day ago
It's slop on top of slop on top of slop. It's not even quality slop. Apart from bloviated self-aggrandising blog posts, the ideas are trivial, and the execution is beyond horrendous.
Look beyond the ChatGPT-generated terminology to see the supervisors, loops and worker processes of Gas Town. Now it's a freelance board with AI agents as freelancers. But sure, polecats, mayors, stamps and character sheets. Whatever floats your upcoming crypto rug pull [1].
This is what he writes: "build stuff really, really fast. So fast that your biggest problem will be ideas." We've yet to see ideas built using the slopcoded monstrosities.
[1] https://pivot-to-ai.com/2026/01/22/steve-yegges-gas-town-vib...
verdverm | a day ago
ns;nt (No Skill; No Taste)
https://blog.kinglycrow.com/no-skill-no-taste/
AstKaj | a day ago
All I can take from this is that you must spend more tokens.
Ronsenshi | a day ago
avaer | a day ago
Is it for people with too much money and time on their hands to flex on Github?
I could at least understand the pitch if there was a crypto scam attached, or if folks were getting paid somehow (which might be an interesting social/AI experiment). But that doesn't seem to be the case.
zachmu | a day ago
rbtprograms | a day ago
cestith | a day ago
verdverm | a day ago
separate company, not aware of any association, dolt has been around for a while
corporat | a day ago
brendoelfrendo | a day ago
tadfisher | a day ago
DragonStrength | a day ago
barrkel | a day ago
drivebyhooting | a day ago
pron | a day ago
npilk | a day ago
egypturnash | a day ago
rektomatic | a day ago
skybrian | a day ago
https://www.dolthub.com/
verdverm | a day ago
skybrian | a day ago
verdverm | a day ago
zachmu | a day ago
verdverm | a day ago
egypturnash | 18 hours ago
zachmu | a day ago
mattnewton | a day ago
zachmu | a day ago
Yegge's Medium uses a serif font so you can tell, but in many faces you can't.
(We still get this comment constantly and it's very unfortunate)
egypturnash | 18 hours ago
corysama | a day ago
block_dagger | a day ago
bambax | a day ago
This reads like a speech from Pete Hegseth.
"Let's do war! Hard! Let's build stuff! Stat!"
Build what? Fight for what?
"The hell if we know! Just get busy dropping bombs, or "stamps" or whatever! Faster!"
In the end, all that's left is, indeed, a wasteland.
alecbz | a day ago
bloggie | a day ago
block_dagger | a day ago
shepherdjerred | a day ago
nkzd | a day ago
pocksuppet | a day ago
verdverm | a day ago
nkzd | a day ago
verdverm | a day ago
https://www.theguardian.com/science/shortcuts/2019/nov/19/do...
plagiarist | a day ago
RGamma | a day ago
juancn | a day ago
Is this Ethereum related?
Help?
senordevnyc | a day ago
And if they don't, so what? Who cares? Why be angry at Steve for playing around with a fun hobby project that goes nowhere?
zachmu | a day ago
ufmace | 23 hours ago
I'm not actually a database engineer with 30 years of experience. If somebody demanded that I pretend to be one, I guess I'd give it a shot, but I would expect any actual employer would be able to tell that I don't have the level of knowledge and experience that you'd expect from somebody like that.
If the base LLM actually has the knowledge of all of these specialties, why can't it just apply them all at once, instead of needing to be told to I guess pretend to be only one of them.
DragonStrength | a day ago
I mean, he can't explain what he's building except pickaxes to make more pickaxes, so it's a bit suspicious. It's just incredible how much impact he has had with this little hustle, given his products are basically turtles all the way down.
g-b-r | 19 hours ago
What impact did he have?
troupo | a day ago
People dislike scam artists, hype artists and bullshitters. Especially when said artists had something actually useful to contribute once upon a time. E.g. Yegge's Platform Rant [1] is still required reading IMO.
Now he's uncritically and unapologetically pushes extremely low quality level AI slop while first trying to prop up Amp, then trying to sell a book, then trying to sell a crypto scam, now trying to sell a vibe-coded database. All the while proclaiming his projects have the basest of basic ideas but somehow need hundreds of thousands of lines of AI-generated low quality slop code to barely function.
The contempt here is the same as for idiots who uncritically run clawdbot and other AI bullshitters and grifters.
Compare this to @simonw who constantly evaluates coding agents, explains what he does in a coherent clear language that doesn't use ChatGPT to invent new inane terms for existing things, and stands behind his work and motivations: https://simonwillison.net
[1] https://gist.github.com/chitchcock/1281611
DragonStrength | a day ago
ozten | a day ago
citrusx | a day ago
You can be part of the new reputation economy, IF you can afford Multiple Claude Max subscriptions and use all your tokens for it.
Kind of like "getting good" at a collectible card game. It's more weighted toward whether you have money (and are willing to spend it) than anything else.
cadamsdotcom | a day ago
First is understanding what to build.
Second is getting the details right.
The best software is both the right software, and high quality. It’s bloody hard to do!
So, we do requirements gathering, and try to get the details right (TDD, continuous delivery etc).. it’s surprising to see Gas Town do none of this and optimistically hope agents will converge on good software by just throwing tokens at a wall and hoping it sticks.
So anyway, that’s what they’re ignoring. What are they actually doing???
It’s all for themselves.
Gas Town’s “product” is the warm fuzzies it gives people with money to burn, warm fuzzy feelings of being “at the frontier”. It’s a luxury product for nerds, and the only ones making money or selling anything are the big labs. There’s zero output or benefit to society because that’s simply not the point.
mieubrisse | 19 hours ago
I think Gastown is truly special, but I wanted something more focused on learning as I think that's the real bottleneck. So I built AgenC to make it trivial to roll learnings back into your Claude.
ebcode | 22 hours ago
planckscnst | 20 hours ago
I've honed a relatively decent flow that requires interaction from me for important parts (mostly) while making its own decisions at the not-important parts (mostly). This results in being able to send the agent off on an hours-long dev cycle and have relatively decent results after that need a few minor fixes. I think this is the best style for the current generation of AI