Auto Generated via Readwise
These are my highlights created/synced via Readwise.io - Referral Link
About
Title: From IDEs to AI Agents With Steve Yegge
Authors: The Pragmatic Engineer
Category:articles
Tags:vibe,llm,ai
Number of Highlights: 14
Source URL: https://www.youtube.com/watch?v=aFsAOu2bgFk
Date: 2026-04-30
Last Highlighted: 2026-04-30
Summary
Steve Yegge says big tech innovation is slowing, but AI is changing software development fast. AI helps both programmers and non-programmers create more and better code. This shift will reshape how companies work and who benefits from software productivity.
Document Note
ONE-SENTENCE SUMMARY:
The conversation argues that AI is rapidly transforming software engineering—moving work up the abstraction ladder, enabling agent orchestration (Gast Town), creating huge productivity gains and new risks (burnout, value capture), and forcing engineers and companies to adapt or be left behind.
SUMMARY:
In this episode Steve Yagi, a veteran engineer and creator of Gast Town, explains eight levels of AI adoption for developers and how agents/orchestrators are changing software development workflows. He argues AI will dramatically increase individual productivity, enable small teams to outcompete big companies, and requires new practices to manage agent coordination, token burn, and the human costs of supercharged work.
FRAMEWORKS & TECHNIQUES:
Eight levels of AI adoption — 00:09
Idea: A spectrum from “no AI” to running multiple agents in parallel describing progressive trust and delegation to AI in engineering workflows.
Quote: “Level one, no AI… At level six, you’re bored because your agent’s busy… It’s agents running agents.”
Agent orchestration (Gast Town) — 08:35
Idea: Build an orchestrator where a single “mayor” coordinates specialized worker agents (minimax “crew” for big-context tasks and short-task “pcats”) to break complex work into parallel subtasks.
Quote: “Gas Town is an orchestrator… one mayor that you talk to, that’s your person, and then whatever else needs to get done, they’re just going to fire off workers.”
Token burn as experiment metric — 01:20
Idea: Measure and encourage token usage (“token burn”) as a proxy for experimentation and learning—higher token burn shows teams are trying and iterating with AI.
Quote: “One, probably the single most important proxy metric that you can have in a company today is token burn because what token burn says is your engineers are trying to do stuff.”
Agent development lifecycle / AC/DC (Guide, Generate, Verify, Repair) — 02:05
Idea: Use a deliberate four-phase loop for AI-generated code: guide (provide canvas/context), generate (produce code), verify (check correctness/security/maintainability), and repair (automated fixes).
Quote: “Guide… Generate… Verify next… any issues identified are provided to a code repair agent to fix.”
Manage “vampiric” productivity and value capture — 01:10
Idea: Recognize AI creates intense bursts of high-value work but drains mental energy; organizations and individuals must set boundaries, redistribute value (equity/comp), and limit hours to sustain engineers.
Quote: “There’s a vampiric effect happening with AI where it gets you excited and you work really, really hard… you might only get three productive hours out of a person at max vibe coding speed.”
QUOTABLE LINES:
- 00:09 “It’s agents running agents.”
- 01:10 “There’s a vampiric effect happening with AI where it gets you excited and you work really, really hard.”
- 08:35 “Gas Town is an orchestrator… one mayor that you talk to, that’s your person, and then whatever else needs to get done, they’re just going to fire off workers.”
Highlights
There’s a vampiric effect happening with AI where it gets you excited, and you work really, really hard, and you’re capturing a ton of value. For me, I’m doing it all for myself, and it’s still kind of like pushing me to my ragged edge.
Note: This was the first time I heard of this and it was really an lightbulb going off.
And what’s happening is, see, companies are set up to extract value from you and then pay you for it.
Right? But the way all companies have always been set up is that they will Give you more work until you break. If you can do it, they’ll just happily just say, “Give you more. I give you more until you your plate flows over and you die.” And people have to learn the art of pushing back, right?
And that’s been a thing for a long time. But it’s changed the equation. The way you push back, the reasons to push back and all that have changed very dramatically and are changing right now because you’ve got all these people who can be super productive. And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
Note: Companies always want the most they can get out of you. But now the vampiric effect is costing you more then time. If we suddenly can output 100x more and we output 100x more at the same comp (with a new mental burden) is that fair?
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And it’s like, let’s say an engineer can be 100 times as productive, just for sake of argument. All right. Who captures that value? If the engineer goes to work and works for eight hours a day and produces 100 times as much, the company captured all of that value. Y And that is not a fair capture exchange.
And so, I, I seriously think founders, and company leaders, and engineering leaders at all levels, all the way down to line managers, you’re going to have to be aware of this and realize that getting your engineers onto this, this treadmill is pulling them into they’re using much, much more of their system 2. You know, they’re doing much more of that hard thinking. Now, the easy stuff is getting automated by.
So, you’re, you’re actually draining them at a higher rate. Their batteries are draining at a higher rate. You might only get three productive hours out of a person at max vibe coding speed, and yet they’re still 100 times as productive as they would have been without AI. So, do you let them work for 3 hours a day? And the answer is, yeah, you better, or your company’s going to break. It’s very interesting because also, like the, the value extraction, I think I, I can see us.
Note: Good example of how we might be in constant System 2 (see Thinking, Fast and Slow) and that we’re drained even faster. And this won’t be sustainable.