The Shape of Things to Come, Part 1: The Continuous Thunderdome — Steve Yegge
Skip to content
Today we're going to explain the "loops and graphs" thing, and I'll show you how<br>to get your coding agent to work all night on massive problems while you snore<br>peacefully.
Then we're going to look at what happens to harnesses once you solve those<br>problems. They either devolve into chaos, or they evolve into cities. Done properly,<br>civilization emerges, right there in your project. We're going to learn the basics,<br>and then we will explore how to build your cities with some taste and class, in a<br>way that makes them good for their citizens to inhabit.
We'll explore a new harness I've been building, called Wheelhouse. It is<br>closed-source, made just for me. I have given up on building reusable<br>harnesses. Indeed I believe harnesses will all soon be bespoke, and the people<br>trying to sell you one will all soon be bebroke. Harnesses need to be part of your<br>application, chemically bonded in. You won't have any luck with someone else's<br>"reusable" harness framework. You don't need it.
Gas Town was intended to be reusable, but I only ever wound up using it to build<br>itself. Gas Town fell apart at the seams with Opus 4.7. Up through 4.6 it was<br>working brilliantly. With 4.7 we saw the introduction of the "just two more things"<br>tic, which prevented Opus from ever converging on being ready to do real work—it<br>always wanted to fiddle with Gas Town itself. The Opus tic never went away, so Gas<br>Town effectively burned down. It had other problems, too, but 4.7 was the final<br>straw.
Since Claude Fable 5 dropped, I have returned full-time to my 30-year-old video<br>game, Wyvern, which I began work on in 1996, and launched in 2001. I had been<br>waiting for a model smart enough to help me with my game, and now that it's here,<br>Wyvern's once again my main squeeze. On the side I do occasional six-figure gigs<br>where I fly to companies and teach them my techniques, and that helps with my<br>(considerable) token bills. But aside from those few paid field trips, I am<br>laser-focused on my game, seven days a week. Wyvern's time will finally come next<br>year. I have officially entered Sam Altman's solo unicorn contest.
Wyvern's React web client, 2026. Rhialto is chatting with Pado's<br>alt char in the pirate town of Stensele.
The game was ultra bit-rotted, despite still running today with paying<br>players. (Once you're hooked on Wyvern, you're hooked for life. I have whales that<br>have spent thousands on it.) I have slowly been using Fable to bring it back to its<br>former glory. I had a 100-year work backlog when I quit Wyvern development again in<br>2022, and Fable has already worked its way through over half of it. Opus could not<br>(and still cannot) understand Wyvern, but Fable wields my code base like a<br>sword.
It's still just a sword, though, and we're trying to build a city. Someone from<br>Anthropic asked me recently what I would do once Fable can write everything for<br>me. On reflection, it was the silliest question I've been asked in many years, and<br>I'm still surprised that someone from Anthropic could think any model could<br>just "write everything."
Making a halfway decent MMO takes many tens of thousands of Fable sessions,<br>throughout which you must display intense focus, dedication, and no small amount of<br>taste. And Fable still can't do the content half: artwork, maps, lore,<br>storylines. All of today's models, including the best image generators, produce GPT<br>3-era slop for all my content, which the players immediately reject. But I can tell<br>you this: once a model arrives with some Wyvern-y taste, we will still have<br>tens of thousands of sessions ahead.
Building large software remains hard. And it always will be, because our ambition<br>will forever outstrip the metal.
With Wheelhouse, I have reinvented something strangely Gas Town shaped from first<br>principles, but it's running many more agents, and they are far more organized. This<br>has shown me the shape of the changes coming next year. I'm operating about 12<br>months in the future. I managed this only through luck: for the past 18 months I<br>have had the magic trio of time, money, and energy, and I've been spending it all<br>on figuring this stuff out. And I am grateful for the privilege.
I did predict much of this post, fifteen months ago, with Revenge of the Junior<br>Developer. It's all going exactly on schedule. I am running fleets of agents,<br>budgets are soaring, all of it.
And that prediction was back when I wasn't really sure what was going on. Now I<br>know exactly what is going on. I know how enterprises will operate, how<br>humans and models will interact. And I am out there building it. I want to be<br>first.
It's important to observe that my work on my game is not special, nor<br>unusually-shaped in any way. It's just work. So all this crap happening to me right<br>now, it will all happen to you soon too. That's how prediction works. Don't be<br>special, stay out in front, and you will see the future clear as day. And then<br>people...