A Return to Code

"Well, yeah, let me talk about vibe coding and how I got into it."

"行,那我就讲讲 vibe coding,以及我是怎么入坑的。"

"So around December of 2025, the coding agents in AI hit an inflection point with the release of Claude Opus 4.5."

"大概在 2025 年 12 月,随着 Claude Opus 4.5 的发布,AI 编码 agent 来到了一个拐点。"

"And people started using it and were like, 'Wow — this is an agent that stays on track, can build apps soup to nuts, can solve thorny problems, and really feels like having a junior programmer at your disposal who's fast, essentially free, and ready to please.'"

"大家用上之后纷纷感叹:'天哪,这个 agent 能盯住目标不跑偏,能从头到尾把 app 全做出来,还能解掉那些棘手问题,真的就像随时听你差遣的初级程序员 —— 快、几乎不要钱,而且一心想讨好你。'"

"That was an inflection point, and I was reading all the hype on Twitter, but this time it felt real."

"那是一个真正的拐点。我在 Twitter 上看到铺天盖地的吹捧,但这一次,真的有东西。"

"I've tried the coding agents in the past with some mixed results, but this time I really got into it. And I haven't seriously coded in decades."

"我以前试过编码 agent,效果参差;但这次我是真的入坑了。要知道我已经几十年没认真写过代码。"

"I have a computer science degree; I understand computer architecture and networking, a little bit of chips, algorithms, et cetera. But I haven't seriously coded in a long time."

"我有计算机科学的学位;计算机体系结构和网络我都懂,芯片、算法之类也略知一二。但我已经很久没认真敲代码了。"

"And the activation energy to writing code is really high. You have to hook up all these different services to each other."

"写代码的'激活能量'实在太高 —— 你得把一堆乱七八糟的服务接到一起。"

"Everything from GitHub to maybe some backend — you're doing Vercel or Firebase or Railway or whatever — and just lots of things to connect together."

"从 GitHub 到后端 —— 不管是 Vercel、Firebase、Railway 还是别的什么 —— 反正一大堆东西要串起来。"

"You have to know lots of jargon — lots of tools. And the AI now makes it really easy."

"你得记一堆术语、一堆工具。而 AI 现在把这件事变得非常容易。"

"So I started with Claude Code like everybody else. I've also used Codex for some of the thornier bug solving and deep problems, and I immediately got addicted. It was incredibly fun."

"我跟大部分人一样,从 Claude Code 开始。一些更棘手的 bug 和深问题我会用 Codex。然后我立刻就上瘾了 —— 太好玩了。"

"And so: what's changed? Well, the agents are really working. These are not just coding assists now — where you ask it to solve a specific problem, it gives you a pile of code, and then you cut and paste that into your IDE, your development environment."

"那么,到底什么变了?agent 是真的能跑通了。它不再只是'编码助手' —— 你提问、它吐一坨代码、你再复制粘贴到 IDE 里那种模式。"

"Rather, you open up a terminal — CLI, as they call it — the command line interface. It's all text-based, which is what these things are really good at, because they're trained on text tokens in the first place."

"取而代之的是,你打开一个 terminal —— 也就是大家说的 CLI、命令行界面。一切都是文本,而文本恰恰是这些模型最擅长的,因为它们本来就是用 text token 训练出来的。"

"It's running Unix inside or underneath. And these agents really know Unix because if you look at all the code out there that they were trained on — sitting on GitHub or elsewhere or Stack Overflow — most of it was Unix."

"底下跑的就是 Unix。这些 agent 对 Unix 烂熟于心,因为 GitHub、Stack Overflow 上能拿来训练的代码,绝大部分都是 Unix 系。"

"And most of the modern OSes are really Unix underneath anyway. macOS is famously BSD."

"再说,大多数现代操作系统底层本来就是 Unix。macOS 众所周知是 BSD。"

"So underneath these are all Unix, which is all text in, text out. So these agents are just long-lived coding AIs that are connected to Unix at a core level."

"所以底层全是 Unix,文本进、文本出。这些 agent 本质上就是长期存活的、内核级地接到 Unix 上的编码 AI。"

"They're connected to the Unix shell so that they can execute commands. They're connected to the file system through basic Unix commands."

"它们连着 Unix shell 来执行命令,通过基本的 Unix 命令访问文件系统。"

"They can call all the Unix commands like grep and awk and sed and pipe and so on — all these operators that daisy chain into each other."

"它们能调用 grep、awk、sed、管道等所有 Unix 命令 —— 这些可以一节一节串起来的操作符。"

"They can run cron jobs so they can be long-lived; and they can spawn more shells and more tasks as needed."

"它们能跑 cron job 来实现长期运行,也能按需开出更多 shell、起更多任务。"

"It's very addictive because — normally, with coding, coding can be really fun once you get into it."

"这玩意儿非常上瘾。其实写代码本来就有意思,一旦进入状态就会很爽。"

"But getting into it, the activation energy is really high. But now all of a sudden you don't have to know all the tools and all the commands."

"问题是进入状态的激活能量太高。可现在,你忽然之间不需要再去记住所有工具和命令了。"

"These things speak English. AIs are incredible translators. And one of their core use cases early on was machine translation."

"这些 AI 说英语。它们是惊人的翻译机器,机器翻译本来就是它们最早的核心用途之一。"

"They were tested on translating. But now they're translating from Python and C and Lisp and Rust, and all of these various programming dialects and all of these specialized commands — and they're communicating in English, and they're very forgiving in their communication."

"它们最早是被用来做语言翻译的;现在则在 Python、C、Lisp、Rust 等等各种编程方言以及各种专门命令之间互译 —— 用英语跟你交流,而且对你的表达非常宽容。"

"So, you can use different words; you can make spelling mistakes; you can explain things your own way."

"所以你可以换词、可以拼错、可以用自己的方式描述。"

"But if you have a basic understanding of computer architecture and networking and programming — and it doesn't take a lot, it can be very basic, actually, very high-level, I should say; not basic in the sense that it's simplistic, but basic in the sense that it's high-level — then you can go very, very far."

"但只要你对计算机体系结构、网络、编程有一点基本理解 —— 不用很多,可以非常基础,确切说是非常'高层级'(不是说简陋,而是说概念级别) —— 你就能走得非常、非常远。"

"And so just for fun, I tried building a bunch of different apps and I started by one-shotting particular apps that I wanted."

"于是我就玩着试着造了一堆不同的 app,从'一次成型'(one-shotting)我想要的那些应用开始。"

"One-shotting meaning: I just give it a description and it gives me back an app. Then I started improving from there."

"所谓 one-shotting,就是我丢一段描述给它,它扔回一个 app。然后我再在这个基础上不断改进。"

"So I actually built my own little app store, which is an app store just for me."

"于是我真的搭了一个属于我自己的'小型 app store' —— 一家专给我一个人开的应用商店。"

"I can ask it for an app; it can deliver that app to my app store, which is a webpage, and eventually I made it into an app itself that lives on my iPhone."

"我可以跟它说我要一个 app,它会把这个 app 送到我的 app store —— 一开始是网页,后来我把它本身做成了一个住在我 iPhone 上的 app。"

"And then I can download those apps with one click, and I can give upgrades like you do with the App Store."

"然后我就可以一键下载这些 app,还能像在 App Store 一样推送更新。"

"So, if I want a new app, for example, that tracks my workouts — and I have this; I built a custom tracking app for just my workouts exactly the way I like it — so I can say:"

"举个例子,我要一个跟踪我健身的新 app —— 我真的有这么一个,完全按我自己的口味做的训练记录 app —— 我会这样描述:"

"So I didn't put all of this in one prompt, but I put a lot of it in one prompt, and I immediately got a working app delivered to my personal app store."

"这些我并没有全塞进同一个 prompt,但相当一部分都在同一个 prompt 里。然后我马上就拿到一个能跑的 app,被送到了我自己的 app store。"

"By the way, the personal app store is a little bit of a joke. It's real in the sense that it's my personal app store: it looks like an app store and my apps get delivered into it."

"对了,'personal app store'多少带点玩笑性质。它真的存在,真的是属于我个人的 app store:长得像一个应用商店,我的 app 会被发布进来。"

"But obviously it's not for wide distribution because Apple gates that. Apple will not let you build apps that can be downloaded on anyone's iPhone."

"但显然它没法面向大众分发,因为这条路是苹果把着的。苹果不允许你做出可以让任何人下载到 iPhone 上的 app。"

"You have to key them against your specific devices. So with my friends and family, I can deliver them apps; I can't yet deliver them to everybody."

"你必须用具体设备来给 app 签名。所以我可以把 app 推给我的家人朋友,但还没办法推给所有人。"

"However, this whole experience is incredibly addictive. You can get extremely customized tuned apps for you."

"不过,整个体验真的极度上瘾。你可以拥有极致个性化、为你专门调过的 app。"

"Now, does this mean that normal apps don't have a place? No, of course they have a place."

"那这是不是意味着普通 app 没位置了?当然不是,它们当然还有位置。"

"Those apps that cover the broad use cases — they're going to be the best-of-breeds."

"那些覆盖通用场景的 app,会是各自领域里的 best-of-breed。"

"Someone's hand-tuned them and slaved over them. So you're not going to beat that if your use case is covered by one of the broad use cases."

"它们是某些人手工打磨、玩命迭代出来的。如果你的需求恰好落在这些通用场景里,你是打不过它们的。"

"But when you want something truly custom or private — these are great for niche apps that only you would want."

"但当你想要的是真正个性化、私密的东西 —— 那种只有你才想用的小众 app —— vibe coding 简直完美。"

"Or when you want to tune them to your specific use case, this is going to be incredible."

"或者当你想把一个工具调到完全契合自己具体场景时,这会带来非常震撼的效果。"

"And it's very addictive — because like in a video game, the way a video game is designed is that it keeps you hooked by giving you feedback and rewards for doing work."

"它非常上瘾 —— 像电子游戏一样。电子游戏的设计就是用反馈和奖励让你保持投入。"

"And it's always at the edge of your capability. So as you get better, the video game gets harder."

"而且它一直把你卡在你能力的边缘。你变强,它就变难。"

"It's not so hard that it's frustrating, but it's not so easy that it's boring."

"不至于难到让你抓狂,也不至于易到让你无聊。"

"So you're always operating at the edge of your capability with a video game and getting these rewards."

"所以玩游戏时,你总是处在能力边缘,持续不断地拿到奖励。"

"But those rewards are fake, and the video game is bounded. It's created by other humans."

"但那些奖励是虚的,游戏本身是有边界的 —— 它是由别的人造出来的。"

"It's sort of a fake little world, and deep down you kind of know that. So you're just figuring out the rules of the game. And then once you've figured out the rules of the game, it's boring."

"它本质上是一个虚构的小世界,你内心其实清楚这一点。你做的事就是摸清规则;一旦摸清,就索然无味了。"

"Except with vibe coding, it's unbounded because now you've got a Turing machine running underneath."

"vibe coding 不一样,它是无界的 —— 因为底下跑的是一台图灵机。"

"You can build anything. The objective is created by you and can keep expanding, so it kind of never fills up completely."

"你想造什么就能造什么。目标是你自己定的、可以一直扩,所以这件事永远填不满。"

"And it has real-world relevance. It's not just some fake world for fake people or fake games that you're solving, so it's way more interesting."

"而且它跟现实世界相连。你不是在解一些虚假世界、虚假游戏里的虚假问题,所以有意思得多。"

"So vibe coding has one-shotted a whole bunch of my friends who have disappeared into vibe coding the apps they've wanted."

"所以 vibe coding 已经一击命中了我一票朋友 —— 他们消失了,埋头去 vibe coding 自己一直想要的 app。"

"But it really, really helps to have a clear direction. You have to know what you want — that's actually the hardest thing — and having a very clear vision of it."

"但前提是你得有明确方向。你得知道自己想要什么 —— 这其实是最难的部分 —— 并且对它有非常清晰的设想。"

"And I have that, because it's a particular app that I was obsessed with for about a year called Airchat — which I built with a team — and it was a social messenger for people to talk through voice and video."

"我恰好有这个东西 —— 一年前我跟一支团队迷恋着做的 Airchat,一款让人通过语音和视频交流的社交通讯产品。"

"It didn't quite work, so we sold it off, got the investors their money back, and got the team some nice packages."

"它没跑通,我们就把它卖掉了,把投资人的钱还了,也给团队拿到了不错的待遇。"

"But I remember that experience as being exhilarating because I was building a product that I wanted and I was working with a brilliant team."

"但那段经历让我兴奋无比 —— 我在做我自己想要的产品,而且和一支极其出色的团队一起。"

"But I had to work through a team to do it. I had eight or nine engineers, depending on the day, and we worked pretty hard for nine to 12 months, and we shipped a couple of variations."

"但我必须通过团队才能做事。每天大概有 8、9 个工程师,我们辛辛苦苦干了 9 到 12 个月,前后发了几版。"

"But with vibe coding, I am basically rebuilding that app. I'm rebuilding from scratch. But the key now is: I'm rebuilding it exactly the way that I want it. There's no compromises."

"现在用 vibe coding,我基本上就是在重做那个 app,从零开始。关键在于:我现在是完全按我自己想要的方式重做它,没有任何妥协。"

"And normally, in the act of building anything with a team, there's always compromises — even if you are not aware of them."

"通常情况下,跟团队一起做任何东西,总会有妥协 —— 哪怕你自己没意识到。"

"Even if you're the dictator in charge, which you rarely are, you still have to just accommodate other people."

"哪怕你是说一不二的独裁者(而你一般并不是),你仍然要照顾别人。"

"You can't say, 'Move this icon left. Now move it right. No, move it back. No, move it back again.' You can't do that. You'll annoy the engineer."

"你没法说:'图标往左移。再往右。不,移回去。不,再移回去。' 你做不出来,工程师会被你逼疯。"

"You can't demand things where you don't have a reasonable justification — where it's just a gut feel or an intuition."

"也不能没理由地下需求 —— 那种只凭直觉、说不清道不明的东西。"

"But the beauty with an AI coding agent is there's none of that."

"但 AI 编码 agent 的妙处在于:这一切都不存在。"

"It's like a self-driving car. You don't feel self-conscious in a self-driving car because there isn't a driver sitting there."

"它就像一辆自动驾驶车。你坐自动驾驶车不会感到尴尬,因为根本没有司机在那儿。"

"The same way, with an autonomous coding agent, you don't feel self-conscious about your own idiosyncrasies."

"同样的,面对一个自主的编码 agent,你不会再为自己那些怪癖而感到拘束。"

"So you can create exactly the thing that you want."

"所以你可以做出完完全全你自己想要的那个东西。"

"I think one of the nice benefits of vibe coding is that — although we may not see like super high-quality code (at least not in this generation), and the architecture needs a lot of work, and these things may have security holes, and they may be hard to scale — the prototyping that you're going to get, the individual apps you're going to get, is going to be very fast and they're going to be true to the vision of the creator."

"我觉得 vibe coding 最棒的好处之一是 —— 虽然这一代写出来的代码品质未必顶尖,架构还需要打磨,可能有安全漏洞、可能不好扩 —— 但你能拿到的原型、你能拿到的个体 app 会非常快,而且会忠实于创作者的设想。"

"There's going to be no compromises."

"不会再有任何妥协。"

"So you may end up with more things like Minecraft — which Notch famously coded by himself — where there was one person's vision."

"你会看到更多像 Minecraft 这样的东西 —— 大家都知道是 Notch 一个人写出来的 —— 完全是一个人的设想。"

"And it may have looked weird because like, 'What is this blocky graphics? It's like a huge step backwards.'"

"它当年看起来甚至很怪:'这种方块画风是什么鬼?简直是技术大倒退。'"

"But he didn't have to compromise. He didn't have to communicate with anybody or explain to anybody why he wanted it that way."

"但他不需要妥协。他不需要跟任何人沟通,也不需要向任何人解释为什么自己要这样做。"

"So I think it expands the scope of discovery."

"我觉得这扩大了'被发现的可能性'的边界。"

"It's also incredibly fun. It takes the number of people who might have built apps from like 0.1 percent to one or two or three percent in the populace."

"它还非常好玩。它把'可能造出 app 的人'从人口的 0.1% 抬到了 1%、2%、3%。"

"Don't get me wrong — the majority of people are not going to code their own apps."

"别误会 —— 大多数人不会自己写 app。"

"For the majority of people, computers are sort of this magic black box and who knows what was going on in there anyway."

"对大多数人来说,电脑就是个魔法黑盒,谁也不知道里面在发生什么。"

"So the fact that it's become 10x or 100x easier still doesn't mean anything to them. It's still a black box."

"所以哪怕它变得容易 10 倍、100 倍,对他们来说也没什么意义,仍然是黑盒。"

"But for the people who are creative, who are self-motivated, and who are articulate and have a good vision, you can code now."

"但对那些有创造力、自驱、表达清晰、有好设想的人来说,你现在能写代码了。"

"There's nobody standing in between you and your prototype."

"你和你的原型之间,再没有任何人挡着你。"

"And yes, if you go to market with a high-functioning app and you need to scale to a lot of users and all of that, then you want to recruit a great team and you want to get real engineers on board, and you're probably going to have to rewrite the whole thing."

"当然,如果你要把一个完整产品推向市场、要扩到大量用户,那你还是要请一支好团队、找真正的工程师上船,而且很可能整套要重写。"

"But if you're experimenting, you're prototyping, you're getting to market, there's nothing better."

"但如果你只是在试、在做原型、在抢着进入市场,没有比这更好的工具了。"

"There's never been a better time to be alive as a creator of software."

"作为软件创作者,从未有过比此刻更好的时代。"

"Now, are the same market opportunities still there? That's a big question. They're shifting very, very fast."

"那原来那些市场机会还在吗?这是一个大问题。它们正在以极快的速度移位。"

"It may be the case that the big companies are vulnerable because now anyone can create software."

"有可能大公司变得脆弱,因为现在谁都能做软件。"

"It may be the case that they have more of an advantage because they have distribution. They can just fill all the gaps with all the software they can dream up."

"也有可能它们反而更占优势,因为它们手上有分发。它们能拿自己想得出来的软件,把所有缝隙全填满。"

"But I actually think this is a renaissance for individual software creators."

"但我倾向于认为,这其实是个体软件创作者的文艺复兴。"

"Now, one other tweet that I put out was something like, 'There's no market for venture-backed software anymore,' or, 'Pure software is not venture investable anymore.'"

"我之前还发过一条推文,大意是:'venture 撑起来的纯软件市场不复存在了',或者说,'纯软件已经不再具备 venture 可投资性。'"

"Yeah, that's a watered-down version of what I really wanted to say, which is that pure software is uninvestable. I would just full stop right there."

"对,那是缓和版。我真正想说的是:纯软件就是不可投资的。我会直接打个句号。"

"If your whole advantage is like, 'Hey, I'm building cool software that other people don't know how to build,' I think that's uninvestable."

"如果你的全部优势是'我能写别人写不出来的酷软件',我觉得那就是不可投。"

"And it's uninvestable for two reasons. One is they can just hack it together today."

"理由有两条。第一,这种东西今天大家都能拼出来。"

"And the second is the coding agents are getting better so quickly that within a year, or even less, they'll probably be building scalable software with good architecture."

"第二,编码 agent 进步实在太快,一年内甚至更短,它们多半就能写出架构良好、可扩展的软件。"

"So I think we're going to see leaps and bounds improvements. That genie is out of the bottle."

"所以我们会看到接连不断的飞跃式提升。这个精灵已经从瓶子里出来了。"

"So if you're a venture investor now, you're looking for hardware, you're looking for network effects, you're looking for AI models."

"所以如果你现在是 VC,你该看的是硬件、网络效应、AI 模型。"

"And I would argue that training AI models is the new building software for however long that lasts until autoresearch and autotraining starts working."

"我甚至会说,训练 AI 模型本身,就是新的'写软件' —— 至少在 autoresearch 和 autotraining 跑通之前都是这样。"

"But I think vibe coding, it's more fun than playing video games. It's more productive. It's more constructive. It has better feedback loops. You build something you want."

"但我觉得 vibe coding 比打游戏好玩。它更高效、更有建设性、反馈循环更好,而且你做的是你想要的东西。"

"You're at the bleeding edge of technology. You may even make some money or career out of it — although careers are kind of dead — but you may make an interesting opportunity out of it."

"你站在技术最前沿。你甚至可能从中赚到钱或拿到一份事业 —— 虽然'事业'这个概念差不多死了 —— 但你确实有可能从中长出一段有意思的机会。"

"And you learn a lot about computers just by doing."

"光是做这件事的过程,你就能学到大量关于计算机的东西。"

"I've seen kids who are vibe coding. It's hard to get kids to program."

"我看到过不少在 vibe coding 的小孩。要让小孩学编程一直是件难事。"

"You can throw Swift Playgrounds and ScratchJr and all of that at them and hope that they pick up coding."

"你可以把 Swift Playgrounds、ScratchJr 这些丢给他们,寄希望他们能学起来。"

"But if you throw vibe coding at them, they're going to get instant feedback and instant rewards."

"但如果你把 vibe coding 丢给他们,他们立刻就能拿到即时反馈和即时奖励。"

"Maybe along the way they'll pick up fundamentals because these things still require some skill to operate."

"在这个过程中他们也会顺手把基本功学了 —— 毕竟这些工具仍然需要一些功底才能玩得转。"

"And in the process of operating them, you'll be forced to figure out the command line; and you'll be forced to figure out how basic computer architecture works; and you'll be forced to figure out concepts like caching, and backing off in a network, and sharing streams, and writing to disk; and latency versus bandwidth trade-offs, et cetera, and all of those things."

"在用它们的过程里,你不得不弄懂命令行;不得不弄懂基础的计算机体系结构;不得不弄懂缓存、网络重试退避、流的共享、写盘、延迟和带宽的取舍等等等等。"

"So you'll be forced to learn some basics of computer algorithms and architecture. And it's just a fun way to go."

"所以你会被推着去学一些计算机算法和体系结构的基础。这是一条很好玩的路径。"

"I've been up late nights, probably spending a couple hours every night — the time that used to go into reading, or doomscrolling, or playing video games — is all now in vibe coding."

"我现在每天熬夜到很晚,差不多每晚两小时 —— 以前用来读书、刷信息流、打游戏的时间 —— 现在全在 vibe coding。"

"In fact, that's why I haven't been active on X recently. I've been completely missing on X because I'm buried in Claude and Codex."

"事实上,这就是为什么我最近在 X 上彻底没声音 —— 因为我整个人都埋在 Claude 和 Codex 里。"

"I'm not sure the analogy applies that well, because AI is jagged intelligence, as they say, where it's incredibly smart at some things and incredibly dumb at others."

"我不太确定这个类比成立。因为 AI 是所谓'锯齿状的智能'(jagged intelligence) —— 它在某些事上聪明得吓人,在另一些事上蠢得离谱。"

"And it's structured very differently than humans in that when you're using Claude, you're using the same AI model — even if you have 10 instances of it running."

"它的结构跟人类完全不同:你用 Claude 时,哪怕同时跑 10 个实例,本质上还是同一个模型。"

"So 10 of them talking to each other doesn't really improve its thinking in the same way that 10 humans talking to each other do, because those humans are trained on 10 different datasets."

"所以 10 个 Claude 互相聊天,远远达不到 10 个人互相聊天的那种效果 —— 因为那 10 个人是被 10 个不同的'数据集'训练出来的。"

"Humans are just inherently very creative and think out of bounds. Whereas the AI agents are trained on the same data distribution. They're literally running the same model."

"人类天然非常有创造力、能跳出边界。而这些 AI agent 训练自同一个数据分布、跑的是同一个模型。"

"It's like 10 people with the same brain and the same dataset talking to each other."

"就像 10 个共用同一个大脑、同一份数据的人在互相对话。"

"Sure, just through thermodynamics they might have some different ideas and come up with something slightly different, but they're generally going to think the same."

"当然,靠热力学波动,他们可能会冒出一些不同的小点子;但整体上他们想得一样。"

"So all you're doing when your 10 agents are talking to each other is you're just throwing 10 times as many tokens at the problem."

"所以你让 10 个 agent 互相聊天,实际上做的事就是把 10 倍的 token 砸到同一个问题上。"

"It's like saying take 10 times as long if you need to."

"等价于说:'如果需要的话,你就花 10 倍时间想。'"

"Now there are different models like Codex, and Gemini, and Grok Code, which are trained slightly differently. Not that different, but they're slightly different. And so they might have some different insights."

"现在的确有 Codex、Gemini、Grok Code 这样的不同模型,它们的训练略有差别。差别不大,但毕竟不同,所以会带来一些不一样的见解。"

"Claude has really good visual presentation through a system called Artifacts and Claude is very good at talking to me at the level that I'm at."

"Claude 通过一个叫 Artifacts 的系统,有非常好的可视化呈现;而且 Claude 很善于在我所处的那个层级跟我对话。"

"So it's very tuned to figure out from your question and your conversation what you're capable of understanding and what level you're asking the question at. It's very good at meeting you at that level."

"它能从你的问题和对话中,精准识别你能理解到什么程度、你是从什么层面问的;然后在那个层面上跟你接住。"

"ChatGPT is still the OG. It's very good all around."

"ChatGPT 仍然是元老,各方面都很均衡。"

"Gemini is very good at search because it has the Google crawl underneath."

"Gemini 搜索很强,因为底下是 Google 的爬虫数据。"

"It's a frustrating product — it's constantly timing out on the app and losing the connection and forgetting the plot."

"它的产品体验让人抓狂 —— 经常超时、断连、丢失上下文。"

"But it's very fast and it's got a great search index. So if the question I'm asking is really a search question underneath, then I use Gemini."

"但它非常快,搜索索引也极强。所以如果我的问题本质上是个搜索问题,我就用 Gemini。"

"Gemini also has access to YouTube. So if you think your answer is lying in a YouTube video — and there's a lot of YouTube videos — then Gemini has the data advantage of YouTube."

"Gemini 还能访问 YouTube。所以如果你觉得答案藏在 YouTube 视频里 —— 而 YouTube 视频实在太多 —— Gemini 就有 YouTube 这份数据的独特优势。"

"So Gemini is really getting by on data advantages. It doesn't feel like the best model to me, but it has the best underlying data."

"所以 Gemini 是靠数据优势在跑。它本身在我感觉里并不是最好的模型,但它的底层数据是最好的。"

"And then Grok is the one I can count on to tell me the truth. It's like the least neutered, least nerfed."

"再说 Grok,这是我能指望它直说真话的那个。它阉割最少、削弱最少。"

"It's got access to X, so it's very good at news. And it's very good at technical problems."

"它能访问 X,所以新闻方面非常强;技术问题上也非常强。"

"So if you're asking a deep, difficult problem in the scientific/mathematical domain, then I think Grok is actually quite good — not that the others aren't, but I just think Grok is standout there."

"所以如果你问的是科学、数学领域里又深又难的问题,我觉得 Grok 其实非常出色 —— 不是说别的不好,但 Grok 在这块尤其突出。"

"And that reflects the biases of the companies that created them and trained them and are driving them."

"这些差异反映的是它们各自背后公司、训练者、推动者的偏好。"

"Currently all four of the leading frontier models have a place."

"在当前这一刻,四个领头的前沿模型都各有一席之地。"

"I do use them against each other. So for example, I wire it up with my GitHub so that every time I'm submitting a new piece of code — say that's written by Claude — then Codex and Gemini automatically fire in every pull request."

"我确实让它们彼此对抗。比如我把 GitHub 串起来:每次我提交一段新代码 —— 比方说是 Claude 写的 —— Codex 和 Gemini 就会在每个 pull request 里自动跳出来 review。"

"It's misnamed, but it's when you actually push code into your main repository and you're basically saying this is ready for review and this is ready to get merged into the main codebase."

"PR 这个名字其实有点歧义,但实际意思是:你把代码推到主仓库,告诉大家'这段已经可以被 review、可以并入主干了'。"

"So you've been working locally in a piece of code, let's say with Claude, and then you push it into the main repository, so you file a pull request. Well, you can set it up so that other agents like Gemini and Codex and Grok automatically fire and review the pull request."

"你在本地写完一段代码 —— 比如用 Claude —— 然后推到主仓库、提一个 pull request。你可以把流程配成:Gemini、Codex、Grok 这些 agent 自动启动来 review 这个 PR。"

"Then they say, 'Oh, well you should change this thing about the architecture,' and so on."

"然后它们会说:'你这个架构应该这样改,'诸如此类。"

"That's a way of getting them to sort of communicate with each other, to have a council — a roundtable of AIs."

"这等于是让它们彼此沟通,组一个'AI 圆桌会议'。"

"But I haven't found that to be as useful as you might think. There's still a lot of groupthink with these AIs."

"但我发现这没你想象中那么有用。这些 AI 之间仍然有大量群体思维。"

"If you're coding with them and you push towards an answer — for example, if you think you know what the answer is — it is rare that they will contradict you."

"在跟它们一起写代码时,如果你朝某个答案推 —— 比如你已经认定了答案 —— 它们极少反对你。"

"You'd have to be pretty wrong for them to contradict you."

"除非你错得离谱,它们才会跟你对着干。"

"They're trying to please you, and I don't think they have any long-lived theory of mind of their own."

"它们一心想讨好你,而且我认为它们并没有真正持久的'自我心智模型'。"

"So they're always kind of morphing towards you, and they're going to find the answer that you are looking for."

"所以它们总是朝你的方向变形,最终给出你要的那个答案。"

"So if you think the answer is in a certain area and you push the models even slightly, all of them will find roughly the same answer because you're leading them to the answer."

"所以你只要心里觉得答案在某个方向,然后稍微推它们一下,它们全都会差不多落在同一个答案上 —— 因为是你在带它们走。"

"They're very easily led around."

"它们极容易被牵着走。"

"One of the things I've noticed is that as the codebase has gotten more complex and larger, it becomes more difficult to manage because it doesn't all fit into the model's context window anymore."

"我观察到的另一件事是:随着代码库越来越复杂、越来越大,管理变得越来越难,因为它已经塞不进模型的 context window 了。"

"The models can only hold a certain amount of data in their heads. And right now the state of the art is about a million tokens, which will be considered laughable in the future."

"模型脑子里只能装下有限的数据,目前最强的大约是一百万 token —— 这个数字以后看会显得可笑。"

"You can approximate that by thinking that is a million words, and that's because of the transformer attention mechanism underneath which, for it to properly work, the problem is a square of the number of tokens in the context."

"你可以粗略想成是一百万词。受限的根源是底下的 transformer attention 机制 —— 它的计算复杂度是 context 中 token 数的平方。"

"So if it's a million tokens, that means the context window is like in the order of complexity of a trillion tokens because it's the square of a million."

"所以一百万 token 的 context window,实际计算量量级在'万亿 token'这个级别 —— 因为是一百万的平方。"

"So the context window runs out as your codebase gets larger. The models can't keep all of it in memory anymore."

"于是,代码库一变大,context window 就告急,模型再也装不下整套代码。"

"So they start making guesses, approximations, they start compacting the context window."

"它们于是开始猜、开始近似、开始压缩 context。"

"They start losing the plot. They get lost. They start fixing the wrong thing."

"它们开始迷路、跑偏、修错地方。"

"They fix the same bug five times."

"同一个 bug 来回修五次。"

"They go do a quick patch in the architecture when the problem lies somewhere else, and you have to guide them."

"问题明明在别处,它们却跑去给架构打个临时补丁。这时你必须出来指挥。"

"So as you are dealing with a more and more complex codebase, it falls upon the operator to provide the guidance to say, 'Actually here, I think we should just re-architect that whole thing.'"

"所以当你处理越来越复杂的代码库时,操作员的指挥就变得不可或缺,你得说:'其实这里,我觉得整块都该重写架构。'"

"And they will do some incredibly boneheaded things."

"它们也会干一些极其愚蠢的事。"

"Like if you are not paying attention and just text is scrolling by, occasionally they'll patch a bug just by eliminating the use case or destroying the feature in the first place."

"如果你不留神、文字哗啦啦滚过去,有时它们会直接把那个 use case 干掉,或把功能本身废掉,就这样'修'好了 bug。"

"Or they'll do something that is clearly a hack and you kind of have to stop them and say, 'Hey, that's a hack.'"

"或者它们会写出一个明显的 hack,而你得叫停它们说:'喂,这是个 hack。'"

"And by the way, I do this all the time."

"顺便说一句,我天天这么做。"

"I'll stop the model. And I'll say, 'No, that's a hack. That's a patch. Go fix it at an architectural level.'"

"我会打断模型,告诉它:'不行,这是 hack,是补丁。请到架构层面去修。'"

"And what's funny is the model will always say, 'Oh, I'm sorry. You're right. That was a hack.'"

"有意思的是,模型一定会说:'对不起,你说得对,那确实是个 hack。'"

"Even if that wasn't a hack, the model will say, 'You're right. That was a hack.'"

"哪怕那根本就不是 hack,模型也会说:'你说得对,那是个 hack。'"

"So the model is always trying to please you, and it doesn't know any better."

"模型永远在讨好你,而且它也分不清。"

"In that sense, it's a little bit like a dog. It's better than you at catching that duck if you're duck hunting with a dog, but it's still a dog."

"从这个角度,它有点像狗。如果你带它去打鸭子,它抓鸭子的本事比你强;但它终究是狗。"

"So if you point it at a bird that's not a duck, it might take that bird down instead."

"如果你指向一只不是鸭子的鸟,它也可能就把那只鸟拿下了。"

"So you do have to guide it. It does require a lot of operational oversight."

"所以你必须指挥它。这件事确实需要大量的操作监督。"

"So, long-winded way of saying, you still have to guide these models."

"绕了一大圈,我想说的是:你仍然必须指挥这些模型。"

"Them talking to each other isn't going to fix the problem."

"让它们彼此聊天解决不了这个问题。"

"And you do have to get involved in the architecture, the debugging, the features, and pay close attention."

"你必须亲自介入架构、调试、功能,并且保持高度专注。"

"But this combo right now of human operator combined with state-of-the-art coding model can yield incredible results."

"但当下,'人类操作员 + 顶级编码模型'的组合,能产出惊人的成果。"

"You can already completely one-shot simple apps. So like a basic task list, a basic video game clone — you can one-shot them: one prompt and you get something that's reasonably good coming out the other end."

"简单 app 你已经可以完全 one-shot 出来了。基础任务清单、基础小游戏克隆 —— 一个 prompt 进去,出来的东西就已经相当不错。"

"So you can see where this is headed."

"所以你能看到这件事将会走向哪里。"

"Eventually, once they have enough data, they will be able to one-shot very complex apps, and that's a whole different world that we're going to get into."

"等数据足够多的那天,它们就能 one-shot 出非常复杂的 app —— 那将是一个完全不同的世界。"

"Now in terms of what is it about coding that makes them uniquely good at it?"

"那么,代码到底有哪些特性,让模型在这件事上格外强?"

"It's just there's tons and tons of data, and when you're training the model, it's very easy to verify, 'Hey, did you do a good job or not?'"

"原因是:数据极其丰富,而且训练时很容易验证'你做得好不好'。"

"Because the code has to compile. It has to execute."

"因为代码必须能编译、能执行。"

"And you can have simple tests that are pre-written on the other side to say, 'Did the code you wrote pass the test? Did it do the thing you're supposed to do?'"

"你可以在另一头写好简单的测试来回答:'你写的代码通过了吗?它做到了它该做的事吗?'"

"So coding turns out to be one of those things that it's actually quite easy to train models on."

"所以代码恰好属于'很容易拿来训练模型'的那一类问题。"

"Mathematics is actually similar in that you have a ton of data — you have a lot of solved problems — and you can verify the output very easily."

"数学其实也是一样:数据极多 —— 大量已解的问题 —— 而且输出极易验证。"

"So in domains where we have a lot of data and you have good verification — self-driving is another one of those. These models do extremely well."

"所以凡是数据多 + 验证好的领域 —— 自动驾驶就是另一个例子 —— 这些模型表现都极好。"

"In areas where you don't have a lot of data, which are brand new fields, the models are not going to do well, and that's still an opportunity for humans and creativity."

"在数据稀少的领域 —— 比如全新的研究方向 —— 模型不会做得好,这仍是人类和创造力的机会。"

"Then domains where it's hard to verify, for example, in creative writing — like who determines what's good creative writing versus what's not, what's slop versus what's not — then these models don't do as well because you can't easily run a closed loop where they're just outputting huge amounts of content and then that content is being immediately algorithmically graded without having to have humans in the loop saying, 'This is good, this is bad.'"

"另一种是难以验证的领域,比如创意写作 —— 谁来判断什么是好的创意写作,什么是 slop —— 这些领域模型就做得不好。因为你没法构造一个闭环,让它们大量产出后被算法直接判分,而完全不靠人类在中间说'这好,那不好'。"

"For example, if you're trying to do creative writing with these models, they're going to output huge amounts of content."

"比如你让这些模型做创意写作,它们能输出海量内容。"

"They can output infinite essays. Who's to say it's good on the other side?"

"它们能写无穷无尽的文章。可另一头谁来说这写得好?"

"Even if you hire some low-wage people to sit around call center style and say, 'this is good' or 'this is bad,' it's only as good as their taste."

"哪怕你雇一帮低薪员工像 call center 那样坐着,挨个判'好'还是'不好',结果质量也只能取决于他们的品味。"

"I think one of the reasons why these coding models got really good recently — there's multiple; one is they're doing sort of almost recursive training where like one model is helping improve the next one — but I think the bigger reason might just be that a lot of the best software engineers started using these models in the last few months and their taste is now feeding back in."

"我觉得这些编码模型最近暴涨,有几个原因:一是它们在做某种近乎递归式的训练,一个模型帮助改进下一个模型 —— 但更大的原因可能是,过去几个月里大量顶级软件工程师开始使用它们,这些人的品味正在反向喂回模型。"

"So you're getting access to their code plus their taste as to what's good and what's not."

"所以模型同时拿到了他们的代码,以及他们关于'什么好什么不好'的判断。"

"You need high-taste feedback loops to improve these models. And those are harder to develop than they look."

"要让这些模型变强,你需要高品味的反馈回路。而这种回路远比看起来难建立。"

"In certain domains it's tractable and in other domains it's hard to see how it happens."

"在某些领域里这条路是可行的;在另一些领域里,我们很难看出它怎么实现。"

"So the obvious stuff is, yeah, you go and you build your app. Great. Less obvious stuff that's like just one level more advanced, which will be laughably simple to a software engineer but it's kind of fun for a non-engineer or someone who hasn't coded in a long time to think about."

"显而易见的玩法是:你去做你的 app,挺好。再进一层,有些事在软件工程师看来简单到可笑,但对一个非工程师、或久未写代码的人来说想起来很好玩。"

"One is I built my own app store. So if I want an app, I literally open up Claude on my phone."

"其中之一就是我做了自己的 app store。如果我想要一个 app,我就在手机上打开 Claude。"

"I can operate a remote terminal, which is running on my desktop, or I can just use Claude in the cloud."

"我可以操控一个跑在我桌面机器上的远程终端,或者直接用云端的 Claude。"

"It can connect to Xcode."

"它能连到 Xcode。"

"I give it a two-line description. It builds me an app. It ships it to my app store. I open my app store app. The app is sitting there. I click install. 30 seconds later, I have a working app on my phone."

"我给它一段两行的描述。它给我做出一个 app,送到我的 app store。我打开我的 app store 应用,app 已经躺在那里。点安装,30 秒后,我手机上就多了一个能跑的 app。"

"That's magical. You can literally be at dinner with someone having a conversation, they describe some app they want, you can describe it to Claude, and five minutes later you're showing them that app on your phone."

"这太魔幻了。你完全可以跟人吃饭聊天时,对方描述一个他想要的 app,你转头描述给 Claude,五分钟后你就能在自己手机上把那个 app 展示给他看。"

"That's why I say it's kind of the beginning of the end for Apple, because Apple relies on their OS and their apps being better than everybody else's."

"所以我说,这有点像苹果末日的开端 —— 苹果依赖的是它的 OS 和它的 app 比别家都好。"

"The hardware, yes, it's better, but it doesn't support their margins and their monopoly, or pseudo-monopoly."

"它的硬件确实更好,但单靠硬件撑不起它的高毛利和它的垄断或类垄断地位。"

"So when all your communication starts going through Claude, or through Codex, or through some other agent, when all you're doing all day long is instead of opening an Uber app, you're saying, 'Call me an Uber,' or instead of opening a workout app, you're saying, 'Where's my workout app? Track my workout. Make no mistakes,' right?"

"所以一旦你所有的交流都开始走 Claude、走 Codex、或别的 agent —— 你一整天里不再是打开 Uber app,而是说'帮我叫一辆 Uber';不再是打开训练 app,而是说'我的训练 app 呢?记录这次训练,别出错,'对吧?"

"Then you are just communicating with the agent, and when that happens, then the need for a phone becomes much smaller and smaller."

"那时你只是在跟 agent 对话,而一旦这件事成立,你对'手机'本身的需求会越变越小。"

"Maybe there's a few banking apps and government apps that haven't ported and don't have the proper APIs. But these agents don't even need APIs. They can figure out and create their own APIs on the fly."

"也许还剩几个银行 app、政府 app 没迁过来、没暴露合适的 API。但这些 agent 根本不需要 API,它们可以现场自己摸出来、造出来。"

"The use case stops being your interfacing with your iPhone or your Android phone. Instead, you're just interfacing with the AI model."

"你的核心使用场景不再是'跟 iPhone 或 Android 手机交互',而是'跟 AI 模型交互'。"

"And now Apple is using Gemini, which is Google's AI model. So what's the difference?"

"而苹果现在用的是 Google 的模型 Gemini。那它跟别家有什么差别?"

"I might as well just use an Android phone, because all I need at that point is I need a screen, I need battery, and I need connectivity. And Android's got that just fine."

"那我不如就用安卓手机 —— 到那一步我需要的只是屏幕、电池和联网,这些安卓都能满足。"

"And then the apps and user interfaces are being created on the fly for what I need."

"然后所有 app 和界面都根据我的具体需求,在当下生成出来。"

"And yes, for certain things, there will always be best-of-breed user interfaces and you'll want some familiarity."

"当然,某些事情上永远会有 best-of-breed 的界面,你也会想要某种熟悉感。"

"But even the era of tap, tap, tap, upgrade your system software, drag this over here, hunt for that button, type into that field, all that is going away."

"但'点点点、升级系统、把这个拖到那里、找那个按钮、在某个输入框里打字'这种时代正在离我们而去。"

"It should all be conversational. It should all be agentic."

"一切都应该是对话式的、agentic 的。"

"And in that world, Apple loses a lot of its advantages, and then it's competing purely on, 'Oh yeah, we have the best chips and we have the best integrated hardware.'"

"在那个世界里,苹果失去大量优势,只剩下'我们有最好的芯片,有最好的整合硬件'可以拿出来打。"

"But that's not the same margins as Apple of today. That's more like the margins that Samsung or Lenovo makes, which is not the margins that Apple wants to have."

"但那种毛利完全不是今天苹果的毛利。那更像三星或者联想的毛利,不是苹果想要的那一档。"

"As a consequence, I think its market cap will compress."

"结果就是,我认为苹果的市值会被压下去。"

"I think Apple giving up on AI will go down as the biggest strategic mistake in the tech industry of this decade, and it's the beginning of the end of Apple's dominance."

"我认为苹果在 AI 上的退让,将被记入这十年里科技行业最大的战略失误,这是苹果统治时代终结的开始。"

"These companies can exist for a long time and make lots of money — like Microsoft is more valuable than it's ever been. But Microsoft Windows has kind of lost the battle because they missed the mobile phone wave."

"这种公司可以活很久、赚很多钱 —— 比如微软现在比历史上任何时候都更值钱。但 Windows 这条线某种意义上已经输掉了,因为它们错过了手机浪潮。"

"They stuck to Windows OS and they didn't upgrade to a touchscreen-based native OS designed for phones from the ground up, and they didn't focus on the consumer."

"它们死守 Windows OS,没去做一个原生为触屏手机设计的全新 OS,也没把重心放在消费者身上。"

"They were too focused at the enterprise level. So Apple surpassed them and is now one of the most valuable companies in the world. I think it used to be the most valuable. It might be Nvidia at this moment."

"它们太专注在企业市场。所以苹果超过了它们,现在是世界上最有价值的公司之一 —— 我记得它曾经是市值第一,这一刻可能是 Nvidia。"

"The same way I think Apple will get surpassed."

"同样的剧本会重演,我认为苹果也会被超过。"

"I think their future growth is capped because they're now captive on AI and they're behind."

"我认为它的未来增长被封顶了,因为它现在在 AI 上被'捕获'了,而且落后。"

"Unless they manage to turn the AI ship around, I think Apple has capped growth long term, and is in 'trouble.'"

"除非它们成功掉转 AI 这艘船,否则苹果在长期意义上增长封顶,我会说它'有麻烦'。"

"Not in the sense that it won't be valuable, but it'll be a lot less valuable than it could have been."

"不是说它不再值钱,而是说它会比它本来该达到的水准,少值很多很多。"

"The other thing is within the app that I'm building, I have a bug reporting infrastructure, where if someone sees a bug, they tap on a button, the bug sends the logs up and the bug files into a server."

"另一件事是,在我正在做的 app 里,我搭了一套 bug 上报基础设施 —— 用户看到 bug,点一下按钮,日志就会上传,bug 进入服务器。"

"And then I have Claude go every 24 hours through all the bug reports and it just fixes them all, by itself, without my having to intervene."

"然后我让 Claude 每 24 小时遍历所有 bug 报告,自己把它们全部修掉,不需要我介入。"

"And it puts all the fixes into side branches for me to review."

"它把所有修复都放进侧分支,等我 review。"

"And then all I have to do is just review the fixes and say, 'Ah, that wasn't really a bug. That wasn't a good fix. Don't ship that.'"

"然后我要做的就是审 review 这些修复:'啊,这其实不是个 bug。这个 fix 不太行,别上。'"

"'Oh, that looks good. Makes sense. Ship it.'"

"'哦,这看着不错,合理,上吧。'"

"I'm just the final gate that decides on what goes out there."

"我只是最后那道门,负责决定什么能放出去。"

"Eventually you can see apps being built that way by features, where the users will ask for features, they'll vote on features, and then there'll be some tastemaker or maintainer in the cloud who'll look at that and say, 'No, the users don't know what they want.'"

"最终你能想象,app 会以功能为单位被这样构建:用户提需求、给功能投票,云端有某个'品味决策者'或维护者站出来说:'不,用户不知道自己想要什么。'"

"Or, 'Oh, that makes a lot of sense. We should fix that or change that.'"

"或者:'这条很合理,应该修一下、改一下。'"

"So I think even software development will become a collaborative process with the users and the agents will be handling all of it."

"所以我相信,连软件开发本身都会变成一个跟用户协作的过程,而具体执行全部由 agent 处理。"

"Because in a sense, the agents can do perfect customer service."

"因为某种意义上,agent 可以做出完美的客服。"

"If your customer service was perfect, your customer service person would also be an incredible coder and would be indefatigable."

"如果你的客服真做到完美,那个客服同时也是了不起的程序员,而且永不疲倦。"

"They would be up 24/7. They would be writing code, fixing bugs, responding to people, and they would have no ego if they wrote a lot of code to fix a bug, and then you just threw it all away."

"它会 24/7 在线,会写代码、修 bug、回复用户;而且即使它为了修一个 bug 写了一大堆代码、最后被你全部丢掉,它也不会有任何自尊心受损。"

"So I just find that kind of a feature very compelling."

"我只是觉得这种'特性'非常迷人。"

"You truly can have one-person, two-person software companies now that can scale to millions upon millions of users and make billions upon billions of dollars."

"你现在真的可以拥有一两个人的软件公司,能扩到数以百万计的用户,赚上亿亿美元的钱。"

"That has happened already in the past with people like Notch and Satoshi Nakamoto, and very small teams like the original Instagram team that just made a huge dent with very few people, or the original WhatsApp team."

"过去当然出现过这种例子 —— Notch、Satoshi Nakamoto,还有最早的 Instagram 团队、最早的 WhatsApp 团队 —— 极少的人就砸出了巨大的坑。"

"But I think you're going to see it more and more now."

"但我相信,从现在起,这种情况会越来越频繁地出现。"