Agentic 代码审查

Coding agents are extraordinarily good now, and getting better fast. The interesting consequence is that the hard part of engineering moved from writing code to deciding whether to trust it, which makes review the most leveraged skill in software right now. How you approach it depends enormously on who you are: a solo developer with no users and a team maintaining a ten-year-old application are not solving the same problem.

如今，编程智能体已经非常出色，并且正在快速进步。一个有趣的结果是，工程中最困难的部分从编写代码转移到了决定是否信任代码，这使得代码审查成为目前软件领域最具杠杆效应的技能。你如何应对这一问题，很大程度上取决于你是谁：一个没有用户的独立开发者，和一个维护着十年老应用的团队，他们解决的并不是同一个问题。

I am more optimistic about agentic engineering than I have ever been. The agents are genuinely good, they get better every month, and on an ordinary day I now ship things I would not have attempted a year ago. This write-up is a map of where the interesting work went, because it did move, and most teams have not fully caught up to where.

我对智能体工程比以往任何时候都更加乐观。智能体确实很棒，它们每个月都在进步，在平凡的一天里，我现在能发布一年前不敢尝试的东西。这篇文章是一张地图，标明了有趣的工作去了哪里，因为它确实转移了，而大多数团队还没有完全跟上步伐。

Code review used to work because of a happy accident of relative speed. A senior engineer could read code faster than a junior could write it, so review kept pace without anyone designing it to, and the team absorbed how the system fit together as a side effect of reading each other’s diffs. A lot of that was not deliberate. It fell out of a single fact: writing code was the slow, expensive part, and reading it was cheap and fast.

代码审查过去之所以有效，是因为相对速度的一个幸运巧合。高级工程师阅读代码的速度比初级工程师编写代码的速度快，所以审查能跟上进度，而没有人刻意这样设计，团队在阅读彼此 diff 的副作用中吸收了系统是如何组合在一起的。其中很多并非有意为之。它源于一个简单的事实：编写代码是缓慢且昂贵的部分，而阅读它是廉价且快速的。

That fact no longer holds. An agent will produce a thousand lines of often solid, well-formatted code in less time than it takes me to read this paragraph, while a human’s reading speed has not changed since roughly the day we started staring at screens for a living. So the constraint moved downstream, to the one step that did not get faster: a person being confident the change is right. I do not think that is a loss. It ...