@dhh: "Now I could write at length about all the caveats. All the potential dangers. All the ways it's not quite there yet. B…
Summary
DHH shares an enthusiastic opinion piece about AI agents, calling them magical and the most fun he's ever had with a computer, heralding an age of endless execution.
View Cached Full Text
Cached at: 08/10/26, 09:36 PM
“Now I could write at length about all the caveats. All the potential dangers. All the ways it’s not quite there yet. But it would be an insult to the marvel of this moment… This is simply the most fun I’ve ever had with a computer.” https://t.co/9TUvyljHHy
Endless execution
Source: https://world.hey.com/dhh/endless-execution-4157e065 The age of agents has brought us endless execution. Every idea, every hunch, every experiment is now within immediate reach. For people with endless ideas, this is nirvana.
It’s also the most fucking sci-fi thing I’ve ever been privy to. The internet was a revelation, but it all felt understandable. The science, the standards, the methods.
But working with agents truly is magic. In the best possible sense of the word. Technology so advanced that it seems scarcely comprehensible. HOW ARE WE MAKING COMPUTERS DO THIS???
I understand the general outline of the underlying mechanism for producing this intelligence: the training, the weights, the math. But the sum effect of those parts is still so wildly, intoxicatingly massive that it’s hard to resist reaching for the metaphysical.
Quantum computing is full of similar mind-bending ideas. Superpositions. The multiverse. All very OUT THERE. But quantum computing isn’t having an impact today. These incomprehensibly clever AI agents are.
And the impact is that I feel like there’s a little genie trapped in my machine now. There to fulfill every wish and every whim. It’s taken months to internalize just what liberation this grants those with the ambition to imagine.
Now I could write at length about all the caveats. All the potential dangers. All the ways it’s not quite there yet. But it would be an insult to the marvel of this moment. AGI is a nefarious concept to pin down, but I’m not sure how different whatever definition we eventually settle on will look from what I’m already experiencing on the daily.
This is simply the most fun I’ve ever had with a computer. And I’ve loved them dearly for over forty years. I loved programming them myself. I loved the games. The configuration. The tactile experience. But none of it as much as I love the power to execute every idea that crosses my mind.
What a time to be alive. Nay, what ablessing.
Similar Articles
@msimoni: I'm not a luddite, but I'm not interested in software created this way. You didn't even _read_ the code, and you want m…
DHH argues that reviewing every line of AI-agent-generated code is unsustainable and that the future lies in adversarial agent reviews plus automated testing, while @msimoni pushes back, expressing distrust of software written by AI that nobody actually read.
DHH went “pencils down.” Is “AI can’t code” still a serious argument?
DHH reveals that 37signals has largely shifted to AI-assisted coding with agents, increasing code output, and questions the ongoing debate about AI's effectiveness in real coding tasks.
Anyone else feel like AI agents are amazing right up until things get complicated?
A reflection on the gap between impressive AI agent demos and dependable real-world execution, arguing that current agents excel at structured tasks but fail under unpredictable conditions, suggesting near-term AI roles will focus on narrow automation with human oversight.
@dabit3: Now that agents can act, we ask: when should they run, what can they touch, how is their work checked, and what context…
The author proposes Automation Engineering as a discipline for designing triggers, guardrails, and success checks to make AI agents safe and reliable without constant human oversight.
@dhh: This has never been more true. The opportunity cost of endless planning and contemplation just went way up. Agents allo…
David Heinemeier Hansson argues that AI agents increase the cost of endless planning and enable rapid experimentation for faster learning.