① 他说过的话

24 条

实验室比我们所有人拥有多得多的 alpha，所以如果他们说这是你应该做的事，那你就应该去做。
the labs have way more alpha than any of us, so if they say this is a thing you should do, then you should go do it.
—— Dex Horthy · [09:30]

指向原始笔记的链接

所以如果你写的代码是你能想到的最聪明的代码、最巧妙的东西，那么按定义，你就不够聪明去调试它。
And so if you write code that is like the smartest code, the most clever thing that you can come up with, you are by definition not smart enough to debug it.
—— Dex Horthy · [14:12]

指向原始笔记的链接

不，不是你的 PR 太多，是你的坏 PR 太多，因为你让人们在写代码的时候把脑子关掉了。
no, you don’t have too many PRs, you have too many bad PRs because you’re letting people turn off their brains uh when they’re coding.
—— Dex Horthy · [18:05]

指向原始笔记的链接

规划全都是关于杠杆的。你怎么为这个任务找到甜点位？
Planning is all about leverage. How do you find the sweet spot for the task?
—— Dex Horthy · [19:52]

指向原始笔记的链接

软件可维护性没有快速的 Oracle（验证器）。
there’s no fast oracle for software maintainability.
—— Dex Horthy · [28:55]

指向原始笔记的链接

如果你在你的软件工厂里交付了坏代码，它会降低未来每一件经过那个工厂的工作的质量，因为模型会看到更多坏模式。
if you ship bad code in your software factory, it degrades every future piece of work that goes through that factory because there’s more bad patterns for the model to see.
—— Dex Horthy · [38:30]

指向原始笔记的链接

你基本上是在从关心你的位置转向关心你的速度。
You’re you’re basically shifting from like caring about your position to caring about your velocity.
—— Dex Horthy · [40:23]

指向原始笔记的链接

审查某样东西大概永远都会有超额收益。我们不知道那是什么。
there will probably always be alpha in reviewing something. We don’t know what it is.
—— Dex Horthy · [43:45]

指向原始笔记的链接

但如果你什么都不审查，那你就只是让 AI 去煮，然后你会得到和其他所有人一样的产品、一样的邮件、一样的公司。
if you don’t review anything, then you’re just letting the AI cook and then you’re gonna get the Same product, the same email, the same company that everybody else gets.
—— Dex Horthy · [43:53]

指向原始笔记的链接

只要我们还在使用基于 transformer 的 LLM，这就是它们工作原理的数学本质，使用更少的上下文你就会得到更好的质量。
As long as we’re using transformer-based LLMs, that’s just like the math of how they work, and you will get better quality if you use less context.
—— Dex Horthy · [52:08]

指向原始笔记的链接

我认为开放的阵营正在以比任何人目前准备注意到的更快的速度追赶上来。
I think that open is catching up faster than anyone is ready to notice yet.
—— Dex Horthy · [55:04]

指向原始笔记的链接

不要抵押你的代码库,指望一年后会出一个聪明到能修复你所有问题的模型。
Do not mortgage your code base in the hopes that in a year the model will come out that’s smart enough to fix all your problems.
—— Dex Horthy · [01:08]

指向原始笔记的链接

这是一个完整的基准测试,专门用来证明模型会让代码库随着时间推移变得越来越糟。
It’s an entire benchmark designed around proving that models make code bases worse over time.
—— Dex Horthy · [02:38]

指向原始笔记的链接

疯狂的事情是这样的:在 Claude 3.5、3.7,大概 Opus 4 那个时代,我们经常这么做——好,是的,我知道代码很潦草,但等到这个问题真正要紧的时候,那就是 GPT-7 的问题了。
Well, so here’s the crazy thing is like we used to do this thing in the like Claude 3.5, 3.7, maybe like Opus 4 days where it’s like, cool, we’re just gonna, yes, I know the code is sloppy, but by the time that matters, that’ll be GPT-7’s problem.
—— Dex Horthy · [06:08]

指向原始笔记的链接

它最终大概会发生,但实验室们似乎对在其他方面爬坡更感兴趣,比如计算机使用和其他前沿领域——软件可维护性那最后 10% 可能和过去三四年我们花功夫的前 90% 一样难,甚至更难。
It will probably eventually happen, but it seems like the labs are much more interested in like hill climbing on other sorts of things like computer use and other like frontiers that like the last 10% of software maintainability may be like as hard or harder than the whole first 90% that we spent the last three or four years on.
—— Dex Horthy · [06:30]

指向原始笔记的链接

你没法捕捉所有东西,而且真相之源应该是代码,而不是某个规格说明或计划。
You can’t capture everything, and the source of truth should be the code, not some specification or plan.
—— Dex Horthy · [09:56]

指向原始笔记的链接

全是炒作。每个人都想互相 impress,每个人都想宣称自己搞明白了 dark factory 那套东西。
It’s all hype. It’s everyone wants to impress each other. Everyone wants to claim they’ve figured out the dark factory thing.
—— Dex Horthy · [18:57]

指向原始笔记的链接

我不认为说「如果你没有用满六个 Codex 订阅,那你就是个糟糕的工程师」是可以的。我觉得这对行业是有伤害的。
I don’t think it’s okay to be like, if you’re not maxing out six Codex subs, then you’re a bad engineer. I think that’s like hurtful to the industry.
—— Dex Horthy · [21:28]

指向原始笔记的链接

从对世界的经济影响来看,我认为每个工程师速度提升 2 到 4 倍对人类社会来说都将是变革性的。
I mean, if you think about it, like in terms of like economic impact in the world, I think every engineer going 2 to 4x faster would be transformational to human society.
—— Dex Horthy · [22:01]

指向原始笔记的链接

但所有人都会错过这一点,因为他们太忙着追求 100 倍了。
But everyone’s going to miss out on that because they’re too busy trying to get to 100x.
—— Dex Horthy · [22:23]

指向原始笔记的链接

我从来没有见过一个提示词或一个确定性工具,能够在没有人类参与的情况下,以任何有意义的方式驱动代码库的可维护性并对抗模型写出不可维护代码的倾向。
But I have never seen a prompt or a deterministic tool that can drive code-based maintainability and fight the model’s tendency to write unmaintainable code in any meaningful way without a human in the loop.
—— Dex Horthy · [27:15]

指向原始笔记的链接

我的建议一直是,不要拿你的代码库去抵押,指望一年后、六个月后或十八个月后会有一个聪明到能解决你所有问题的模型发布出来。
And my advice is always like, do not mortgage your code base in the hopes that in a year or six months or 18 months a model will come out that’s smart enough to fix all your problems.
—— Dex Horthy · [30:27]

指向原始笔记的链接

到那个时候,我是在做你工作中每一个重要的部分。
And at that point, I’m doing every part of your job that matters.
—— Dex Horthy · [32:03]

指向原始笔记的链接

如果你每一千行就检查一次,交付一万行代码会容易得多。
Because like, it’s a lot easier to ship 10,000 lines of code if you check it every thousand lines.
—— Dex Horthy · [39:31]

指向原始笔记的链接

② 出现在这些集

2 集

• 《软件工厂的教训：不读代码的四五个月后，我们重写了整个产品》 — 作为联合主持
• 《「上下文工程」之父 Dex Horthy：智能体编程的真相与工作流》 — 作为主持

③ 他谈到的

点进去有真内容 —— 本页主要出口

智能体 · 上下文工程 · 代码审查 · Claude · Simon Mayfor · HumanLayer · 软件工厂 · Slop Code Bench · 规范 · 暗工厂

④ 也在聊「AI 编程」的人