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

Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

maggieappleton.com

Post reply on HN
Print view View original

351–360 of 466 posts

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#351

Post by antonvs »

potatolicious wrote: post #221
hyperpape wrote: post #97

Earlier quoted context omitted.

> If you read Steve's writeup, it's clear that this is a big fun experiment: So, Steve has the big scary "YOU WILL DIE" statements in there, but he also has this: > I went ahead and built what’s next. First I predicted it, back in March, in Revenge of the Junior Developer. I predicted someone would lash the Claude Code camels together into chariots, and that is exactly what I’ve done with Gas Town. I’ve tamed them to…

> "If you say two different and contradictory things, and do not very explicitly resolve them, and say which one is the final answer, you will get blamed for both things you said, and you will not be entitled to complain about it, because you did it to yourself." If I can be a bit bold and observe that this tic is also a very old rhetorical trick you see in our industry. Call it Schrodinger's Modest Proposal if you w…

It's a version of a motte and bailey argument (named after a medieval castle defense system):

> "...philosopher Nicholas Shackel coined the term 'motte-and-bailey' to describe the rhetorical strategy in which a debater retreats to an uncontroversial claim when challenged on a controversial one."

-- https://heterodoxacademy.org/blog/the-motte-and-the-bailey-a...

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#352

Post by TeMPOraL »

pydry wrote: post #77
mediaman wrote: post #46

I don't get the widespread hatred of Gas Town. If you read Steve's writeup, it's clear that this is a big fun experiment. It pushes and crosses boundaries, it is a mixture of technology and art, it is provocative. It takes stochastic neural nets and mashes them together in bizarre ways to see if anything coherent comes out the other end. And the reaction is a bunch of Very Serious Engineers who cross their arms and h…

>I often feel like our industry has lost its sense of whimsy and experimentation from the early days, when people tried weird things to see what would work and what wouldn't. Remember the days when people experimented with and talked about things that werent LLMs? I used to go to a lot of industry events and I really enjoyed hearing about the diversity of different things people worked on both as a hobby and at work.…

Well, LLMs are an engineering breakthrough of the degree somewhere between the Internet and electricity, in terms of how general-purpose and broadly-applicable they are. Much like them, LLMs have the potential to be useful in just about everything people do, so it's no surprise they've dominated the conversation - just like electricity and the Internet did, back in their heyday.

(And similar to the two, I expect many of the initial ideas for LLM application to be bad, perhaps obviously stupid in hindsight. But enough of them will work to make LLMs become a lasting thing in every aspect of people's lives - again, just like electricity and the Internet did).

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#353

Post by meowface »

wahnfrieden wrote: post #322
wahnfrieden wrote: post #246

Earlier quoted context omitted.

Complete with a "Let’s goooooooo!" And FOMO stories about missing out on Bitcoin when he knew about it, so he doesn't want you to miss out on this new opportunity to get "filthy rich" as an "investor" while you still can.

More details on the pump and dump scheme he joined in on promoting and drew money from: https://pivot-to-ai.com/2026/01/22/steve-yegges-gas-town-vib...

This initially sunk my heart, but in all his replies there are like 50 very clearly unintelligent crypto grifters telling him he needs to be killed for scamming them, so I am unsure who to root for at this point. It's depressing he accepted it, but I might partially forgive it due to him making a lot of them lose money.

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#354

Post by meowface »

JasonADrury wrote: post #348
wahnfrieden wrote: post #322

Earlier quoted context omitted.

More details on the pump and dump scheme he joined in on promoting and drew money from: https://pivot-to-ai.com/2026/01/22/steve-yegges-gas-town-vib...

[flagged]

Because you'd be aiding and abetting a pennystock scam.

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#355

Post by 63stack »

edg5000 wrote: post #343
wordswords2 wrote: post #339

There is nothing professional, analytical or scientific about Gas Town at all. He is just making up a fantasy world where his elves run in specific patterns to please him. There is no metrics or statistics on code quality, bugs produced, feature requirements met.. or anything. Just a gigantic wank session really.

Are you being sarcastic or serious? Meeting requirements is implicitly part of any task. Quality/quantification will be embedded in the tasks (e.g. X must be Y ); code style and quality guidelines are probably there somewhere in his tasks templates. Implicitly, explicit portions of tasks will be covered by testing. I do think it's overly complex though; but it's a novel concept.

Everything you said is also done for regular non-ai development, OP is saying there is no way to compare the two (or even compare version x of gas town to version y of gas town) because there are 0 statistics or metrics on what gas town produces.

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#356

Post by rulelet »

mediaman wrote: post #46

I don't get the widespread hatred of Gas Town. If you read Steve's writeup, it's clear that this is a big fun experiment. It pushes and crosses boundaries, it is a mixture of technology and art, it is provocative. It takes stochastic neural nets and mashes them together in bizarre ways to see if anything coherent comes out the other end. And the reaction is a bunch of Very Serious Engineers who cross their arms and h…

We have a different take than Gastown. If AI behaves unreliably and unpredictably, maybe the problem is the ask. So we looked at backend code and decided it was time to bring in more declarative programming. We are already halfway there with declarative frontend (React) and declarative database (SQL). Functional programming is an answer, but functional programming didnt replace object oriented programming because of practical reasons.

So even if the super serious engineers are serious, they should watch their back. Eventually enough guardrails will be created or even the ask will change enough for a lot of automation to happen. And make no mistake, it is automation no different than having automated testing replace armies of manual testing or code generation or procedural generation or any other machine method. And who is going to be left with jobs? People who embrace the change, not people who lament for the good old days or who can't adapt.

Sucks but just how the world works. Sit on the bleeding edge or be burned. Yes there is an "enough" but I suspect enough is around people willing to look at Gastown or even make their own Gastown, not the other side.

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#357

Post by DonHopkins »

mediaman wrote: post #46

I don't get the widespread hatred of Gas Town. If you read Steve's writeup, it's clear that this is a big fun experiment. It pushes and crosses boundaries, it is a mixture of technology and art, it is provocative. It takes stochastic neural nets and mashes them together in bizarre ways to see if anything coherent comes out the other end. And the reaction is a bunch of Very Serious Engineers who cross their arms and h…

Hi mediaman! I'm totally there with you and Steve on the whimsy and experimentation! And your tolerant attitude gives me the Dutch courage to post this.

I've been reading Yegge since the "Stevey's Drunken Blog Rants™" days -- his rantings on Lisp, Emacs, and the Eval Empire shaped how I approach programming. His pro-LLM-coding rants were direct inspiration for my own work on MOOLLM. The guy has my deep respect, and I'm intrigued by his recent work on Sourcegraph and Gas Town.

Gas Town and MOOLLM are siblings from that same Eval Empire -- both oriented along the Axis of Eval, both transgressively treating LLMs as universal interpreters. MOOLLM immanentizes Eval Incarnate -- https://github.com/SimHacker/moollm/blob/main/designs/eval/E... -- where skills are programs, the LLM is eval(), and play is but the first step of the "Play Learn Lift" methodology: https://github.com/SimHacker/moollm/tree/main/skills/play-le....

The difference is resource constraints. Yegge has token abundance; I'm paying out of pocket. So where Gas Town explores "what if tokens were free?" (20-30 Claude instances overnight), MOOLLM explores "what if every token mattered?" Many agents, many turns, one LLM call.

To address wordswords2's concern about "no metrics or statistics" -- I agree that's a gap in Gas Town. MOOLLM makes falsifiable claims with receipts. Last night I ran an Amsterdam Fluxx Marathon stress test: 116+ turns, 4 characters (120+ character-turns per LLM call), complex social dynamics on top of dynamic rule-changing game mechanics. Rubric-scored 94/100. The run files exist. Anyone can audit.

qcnguy's critique ("same thing multiplied by ten thousand") is exactly the kind of specific feedback that helps systems improve. I wrote a detailed analysis comparing the two approaches -- intellectual lineage (Self, Minsky's K-lines, The Sims, LambdaMOO), the "vibecoded" problem (MOOLLM is LLM-generated but rigorously iterated, not ship-and-hope), and why "carrier pigeon" IPC architecture is a dark pattern when LLMs can simulate many agents at the speed of light.

an0malous raises a real fear about bosses thinking "throw agents at it" replaces engineering. Both systems agree: design becomes the bottleneck. Gas Town says "keep the engine fed with more plans." MOOLLM says "design IS the point -- make it richer." Different answers, same problem.

lowbloodsugar mentions building a "proper, robust, engineering version" -- I'd love to compare notes. csallen is right that "future" doesn't mean "production-grade today."

Analysis: https://github.com/SimHacker/moollm/blob/main/designs/GASTOW...

MOOLLM repo: https://github.com/SimHacker/moollm

Happy to discuss tradeoffs or hear where my claims don't hold up. Falsifiable criticism welcome -- that's how systems improve.

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#358

Post by vanderZwan »

mediaman wrote: post #46

I don't get the widespread hatred of Gas Town. If you read Steve's writeup, it's clear that this is a big fun experiment. It pushes and crosses boundaries, it is a mixture of technology and art, it is provocative. It takes stochastic neural nets and mashes them together in bizarre ways to see if anything coherent comes out the other end. And the reaction is a bunch of Very Serious Engineers who cross their arms and h…

> I don't get the widespread hatred of Gas Town. If you read Steve's writeup, it's clear that this is a big fun experiment. It pushes and crosses boundaries, it is a mixture of technology and art, it is provocative.

Because I actually have an arts degree and I know the equivalent of a con artist in a rich people arts gallery bullshitting their way into money when I see one.

And the "pushing and crossing boundaries" argument has been abused as a pathetic defense to hide behind shallowness in the art world for longer than anyone in this discussion board has been alive. It's not provocative when it's utterly predictable, and in this case the "art" is "take the most absurd parody of AI culture and play it straight". Gee whiz how "creative" and "provocative".

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#359

Post by pydry »

edg5000 wrote: post #343
wordswords2 wrote: post #339

There is nothing professional, analytical or scientific about Gas Town at all. He is just making up a fantasy world where his elves run in specific patterns to please him. There is no metrics or statistics on code quality, bugs produced, feature requirements met.. or anything. Just a gigantic wank session really.

Are you being sarcastic or serious? Meeting requirements is implicitly part of any task. Quality/quantification will be embedded in the tasks (e.g. X must be Y ); code style and quality guidelines are probably there somewhere in his tasks templates. Implicitly, explicit portions of tasks will be covered by testing. I do think it's overly complex though; but it's a novel concept.

>Are you being sarcastic or serious?

I think if you'd read the article through you'd know they were serious coz Yegge all but admits this himself.

Re: Gas Town's agent patterns, design bottlenecks, and vibecoding at scale

#360

Post by DonHopkins »

mediaman wrote: post #46

I don't get the widespread hatred of Gas Town. If you read Steve's writeup, it's clear that this is a big fun experiment. It pushes and crosses boundaries, it is a mixture of technology and art, it is provocative. It takes stochastic neural nets and mashes them together in bizarre ways to see if anything coherent comes out the other end. And the reaction is a bunch of Very Serious Engineers who cross their arms and h…

Hey, I didn't get a harumph out of that VSE crossing his arms at me!

https://www.youtube.com/watch?v=g2Bp8SqYrnE

Post reply on HN