Reflecting on a Year of Claude Code

"When we first released Claude Code, it was just a little video, and I remember posting it to Slack."

"我们第一次发布 Claude Code 的时候,它就是个小视频。我记得当时把它发到了 Slack 上。"

"There was like two people that gave a reaction — people were excited."

"大概就两个人给了个表情回应 —— 显得挺兴奋的。"

"I thought it was really cool, especially for my very easy engineering tasks. It was quite good at it."

"我当时觉得它真的很酷,尤其是处理我那些特别简单的工程任务时,它做得相当不错。"

"That's like a really nice way to say that it wasn't really good. [laughs]"

"你这是在用一种很客气的说法表达'它当时其实不怎么样'。(笑)"

"I can't believe it's only been a year since we first launched Claude Code. It's hard to remember what that was like — it's so different than what we're doing today."

"真不敢相信,距离我们第一次发布 Claude Code 才过去一年。已经很难回想起当时是什么样了 —— 跟我们今天在做的事完全不同。"

"Now I just have like armies of agents that are doing stuff. I'm prompting one agent, or I have an agent that's prompting agents that's prompting agents, and it's like a tree of thousands of agents."

"现在我手里就是一支支 agent 大军在干活。我要么在给一个 agent 下指令,要么是有一个 agent 在指挥别的 agent、那个再指挥下一层 —— 像一棵有成千上万个 agent 的树。"

"But I think it's just like the most important idea when working on this stuff is: every single time Claude makes a mistake, I don't tell Claude to do it differently. I tell it to write it to the CLAUDE.md, or to make a skill, or something to do it differently."

"但我觉得,做这件事最重要的一个想法是:每次 Claude 犯错,我不会去让 Claude '这次换个做法',而是让它把这条教训写进 CLAUDE.md、或者做成一个 skill 之类的东西,从此换个做法。"

"And if you can do this, then Claude can just like run forever."

"只要你能做到这一点,Claude 就能一直跑下去。"

"And I think the other thing that we kind of realized is the verification is really important. We didn't realize that."

"我们后来还意识到另一件事:验证(verification)真的非常重要 —— 这一点我们一开始没想到。"

"I hear this come up a lot with developers and enterprises that we meet with. What are your tips for making Claude Code really good at verification?"

"我们见的开发者和企业里,这个问题经常被提起。要让 Claude Code 真正擅长'验证',你有什么诀窍?"

"I sort of feel like this is this thing that just like everyone misunderstands."

"我感觉这是一件几乎所有人都会误解的事。"

"Whenever we talk about verification, people are thinking like unit test, or they're thinking like lint or type check. These are the things that are obviously really easy to automate, and these are the things that were already automated."

"每次一聊'验证',大家想到的都是单元测试,或者 lint、type check。可这些本来就特别容易自动化,而且早就被自动化掉了。"

"But actually when we talk about verification for agents, it's something slightly different. It's like: can the agent run the thing?"

"但当我们谈的是给 agent 用的验证时,含义其实略有不同。它问的是:agent 能不能把这东西真正跑起来?"

"It takes a little bit of mental work to figure out how exactly you do this, because it's often not straightforward. And I think that's one of the challenges."

"要弄清楚到底怎么做到这一点,得费点脑子,因为它往往并不直接。我觉得这正是难点之一。"

"I remember with Opus 4, Claude tested itself. We just like hooked it up to Opus 4, and I was like, 'Claude, build a feature and then test yourself in bash,' and it opened a little Claude CLI and tested its own feature, and I was just like, whoa, it's crazy."

"我记得 Opus 4 那会儿,Claude 自己测了自己。我们就把它接到 Opus 4 上,我说:'Claude,做一个功能,然后在 bash 里自己测一下',结果它打开了一个小小的 Claude CLI,测了它自己写的功能。我当时就是:哇,太疯狂了。"

"Now we're so used to it. Now we have these loops going for the iOS simulator and the Android simulator and like computers for desktop. It's not surprising, but back then that was crazy. How are you doing it?"

"现在我们已经习以为常了。现在我们有各种循环在跑 —— iOS 模拟器、Android 模拟器、桌面端用计算机操作。如今这一点都不稀奇,但当时真的很炸裂。你现在是怎么做的?"

"So I've been mainly hacking on the desktop app these days, and one of the engineers on the team actually added this desktop development skill that teaches Claude how to run the local desktop app, and I've been having it use it."

"我最近主要在捣鼓桌面端 app。团队里一位工程师做了一个'桌面开发 skill',教 Claude 怎么运行本地的桌面 app,我就一直让它用这个 skill。"

"It still runs into issues or like bugs with the staging environment sometimes. And so what I have it do is, in those cases, I have it read Slack and understand — hey, is staging down right now? Or has someone else already hit this?"

"它有时还是会碰到问题,或者 staging 环境里的 bug。这种情况下,我让它去读 Slack、搞清楚状况 —— 比如:staging 现在是不是挂了?是不是已经有别人踩到这个坑了?"

"And then when it debugs the whole issue, I tell it to update the desktop development skill."

"等它把整个问题调试清楚之后,我会让它去更新那个桌面开发 skill。"

"What the skill does is Claude actually spins up a local desktop app, and it uses computer use to click around it. And so when I add a new UX, it clicks around to invoke the new UX."

"这个 skill 做的事,是让 Claude 真的启动一个本地桌面 app,并用'计算机操作'(computer use)在里面点来点去。所以每当我加一个新的交互界面,它就会点进去、把这个新界面跑一遍。"

"It also tests edge cases, and when there's an issue, it fixes it, ships it, and rechecks."

"它还会测各种边缘情况;一旦发现问题,就动手修、提交、再复查一遍。"

"This is like honestly one of my favorite things about this team — everyone codes. I have never been on a team where like my PM would code, and it's crazy, and your code is like really good."

"说实话,这是我最喜欢这个团队的一点 —— 人人都写代码。我从没待过一个 PM 也写代码的团队,这太疯狂了,而且你的代码写得真的很好。"

"But I also just feel like it's becoming easier, because it's like essentially Claude writes the code. And so what matters a little more is what's the idea that you have."

"但我也觉得,这件事正变得越来越容易,因为说到底,代码是 Claude 在写。于是更要紧的反而是:你脑子里有什么样的想法。"

"And I feel like if you're a person that has like the product context and the business context, and you're thinking about the design and the user, you're just going to come up with better ideas."

"我觉得,如果你是一个既懂产品脉络、又懂业务背景,同时还在认真琢磨设计和用户的人,你自然就会想出更好的点子。"

"It's kind of like all the roles are merging."

"这有点像是 —— 所有的岗位都在融合。"

"I remember seeing Megan our designer's PRs, and I was just horrified at the beginning. I was like, 'Oh my god, why is Megan putting up PRs?'"

"我还记得第一次看到我们设计师 Megan 提的 PR,一开始简直被吓到了。我想:'天哪,Megan 怎么会在提 PR?'"

"And then she was like, 'Yeah, yeah, I'm just like I'm fixing the button.' And I was like, 'Okay, all right, well, the code looks good, so maybe it's fine.' And now it's just like it's totally normal."

"然后她说:'对啊对啊,我就是在修那个按钮。'我心想:'好吧,行,代码看起来没问题,那大概没事。'到现在,这已经完全是家常便饭了。"

"Yeah, we see this across all the enterprises we talk with. The engineers adopt Claude Code first, and then the adjacent roles look over their shoulder and they're like, 'Whoa, this thing is very powerful, let me try it out.'"

"是啊,我们聊过的所有企业里都能看到这一幕。先是工程师用上 Claude Code,然后旁边的岗位探头一看,心想:'哇,这东西太强了,我也来试试。'"

"And we found — it's crazy — we found that our designers are more productive making prototypes and making changes directly in the app instead of paying an engineer."

"我们发现 —— 这点很离谱 —— 我们的设计师如今做原型、直接在 app 里改东西,效率更高了,根本不用再去'雇'一个工程师帮忙。"

"PMs are making changes in the app. Our finance team runs in Claude Code — they do their projections there."

"PM 们直接在 app 里做改动。我们的财务团队就泡在 Claude Code 里 —— 他们的财务测算都在那儿做。"

"Data science, too — like if you talk with our data scientists, it's so cool. It's just like everyone just has Claude Code up on their screens."

"数据科学也一样 —— 你要是跟我们的数据科学家聊聊,会觉得太酷了。基本上每个人的屏幕上都开着 Claude Code。"

"I feel like it's remarkably versatile for different roles. What do you feel like nowadays are the use cases that are pushing the limit?"

"我觉得它对不同岗位的适配性强得惊人。你觉得现在哪些用法是真正在突破边界的?"

"One that I'm super excited about is routines. There's one engineer on our team who launched voice mode across all of our products, and he has this routine set up that just listens for every ticket that comes — every GitHub issue, every bug report about voice mode — and his Claude just picks it up proactively, puts up a fix, and then pings the PR to him."

"我特别兴奋的一个,是 routines(例行任务)。我们团队有位工程师,在我们所有产品里都上线了语音模式。他配了一个 routine,专门盯着每一条进来的工单 —— 每一个 GitHub issue、每一条关于语音模式的 bug 反馈 —— 他的 Claude 会主动接手、提一个修复,然后把 PR 推给他。"

"And when he got that working for voice mode, he thought, 'Okay, we're getting a lot of other feedback that isn't being responded to.' So he also set up a routine to listen for that."

"等他把语音模式这套跑通后,他想:'好家伙,我们还收到一大堆没人回应的其它反馈。'于是他又配了一个 routine 去盯这些。"

"So I shipped this small feature, and there was like an edge case in it that I didn't see. And so someone filed a bug for it, and I was going to get to the bug that night."

"有一次我上线了一个小功能,里面有个我没注意到的边缘情况。于是有人提了个 bug,我本来打算当晚去处理。"

"And as my Claude was working, it said, 'Wait a second, another Claude has already fixed this.' And I was like, 'How's this possible?' I've never talked to him about this feature before."

"结果我的 Claude 干着干着,冒出一句:'等等,已经有另一个 Claude 把这个修好了。'我当时就懵了:'这怎么可能?'我从来没跟他聊过这个功能。"

"And so I pinged him and I was like, 'How did you fix this so quickly?' And he said he has another routine that just looks for bug reports that haven't been responded to in five hours and puts up a fix, and he merges the ones that are easy to verify."

"于是我去戳他:'你怎么修得这么快?'他说他还有另一个 routine,专门找那些超过五小时没人回应的 bug 反馈、自动提交修复;其中那些容易验证的,他直接合并。"

"Claude tells me this like all the time now — that someone else has already fixed it. There's always like another person's Claude that's working on it."

"现在 Claude 老是这么跟我说 —— 说有别人已经把它修好了。总有另一个人的 Claude 正在做这件事。"

"Yeah, that's been one of the changes. A while ago we were trying to figure out how to use routines. The agent SDK was this first idea that we could use Claude Code programmatically, but at the beginning it just wasn't obvious — how do we use it? What do we use it for?"

"对,这是其中一个变化。前阵子我们还在琢磨 routines 到底怎么用。Agent SDK 是第一个让我们意识到'可以用程序化的方式调用 Claude Code'的想法,但一开始并不明朗 —— 怎么用?用来干嘛?"

"And I think routines are the first really obvious application. It just does like all the code review. It babysits like every PR."

"我觉得 routines 是第一个真正显而易见的应用场景。它把所有的 code review 都做了,还盯着每一个 PR。"

"Remember back in the day, you used to actually have to like respond to code review comments. You used to have to fix CI. You used to have to rebase. Like, I haven't done that in a long time."

"还记得以前吗,你得亲自去回 code review 的评论、得去修 CI、得做 rebase。这些事我已经很久没干过了。"

"When you're in the CLI and you're synchronously working with Claude, what are your go-to features?"

"当你在 CLI 里、和 Claude 同步协作时,你最常用的功能是哪些?"

"Okay. What they used to be is plan mode. I don't use that anymore."

"好。以前是 plan mode(规划模式),但我已经不用它了。"

"What do you use instead?"

"那你改用什么?"

"Auto mode."

"Auto mode(自动模式)。"

"It's the best. Instead of plan mode."

"它是最棒的。直接取代了 plan mode。"

"Instead of plan mode, yeah. Because the newer models, they don't actually need like a planning step anymore."

"对,取代 plan mode。因为更新的模型,其实已经不再需要单独的'规划'这一步了。"

"I think this was really important for like Opus 4 through maybe 4.5, then I think starting with 4.6 and definitely with 4.7, it just doesn't need that planning step. I think some people still use it — they like to have that artifact. I don't use it."

"我觉得这一步在 Opus 4 到大概 4.5 那段还很关键;但从 4.6 开始、到 4.7 就更明显了,它根本不需要那个规划步骤。有些人还在用,他们喜欢有那么一份产出物;我不用。"

"I just do auto mode for everything, because then I start my Claude, it starts to work, and then I just like move on to the next Claude, and I don't have to sit there and watch it."

"我什么都用 auto mode,因为这样我起一个 Claude,它就开始干活,我直接转去开下一个 Claude,不用坐在那儿盯着它看。"

"But from the very early stage we had this like permission prompt model for Claude Code, right? It runs a tool, and then it asks you, like, 'Hey, are you okay running this tool?' and you had to say yes or no."

"不过从最早期开始,Claude Code 就有一套'权限确认'的模式,对吧?它要运行一个工具,就会问你一句:'嘿,这个工具可以跑吗?'你得回答 yes 或 no。"

"And at the time that was kind of the best we had a year and a half ago, because we didn't have classifiers, the model was not as well aligned as it is today."

"在一年半前,那已经是我们当时能做到的最好方案了,因为我们没有分类器,模型的对齐程度也不像今天这么好。"

"So auto mode was just such a big step up, because actually you don't want to read most of these requests. Just routing it to a different model and having it check for security works so much better."

"所以 auto mode 是一次巨大的跃升,因为说实话,这些请求里绝大多数你根本不想去读。把它转交给另一个模型、让那个模型来做安全检查,效果好太多了。"

"Yeah. And if a thing is a little sus, or it's not a command that you think you want to run, or it's not safe, the model will just deny it, and then you can go back and you can allow it later."

"对。如果某件事有点可疑,或者不是你想跑的命令,又或者不安全,模型会直接拒掉;之后你可以再回头手动放行。"

"I think this has been one of those like step changes. There's no way we could have done this a year and a half ago."

"我觉得这是那种'阶跃式'的变化之一。这在一年半前我们是根本做不到的。"

"It's just human nature — when you accept 99% of requests, your eyes just glaze over when you read it."

"这就是人性 —— 当你对 99% 的请求都点'同意',读的时候眼睛自然就放空了。"

"And so actually we feel that auto mode is more safe than reading every single permission prompt, because it means that you're only paying attention to the most important thing, and not like being spammed a bunch of things that are just 99% yes."

"所以我们其实认为,auto mode 比逐条去读每一个权限确认更安全 —— 因为这意味着你只把注意力放在最重要的那件事上,而不是被一堆 99% 都该点'是'的东西刷屏。"

"I think security is one of these things — like you can talk about it, and then it's a totally different thing to actually do it correctly, because it just doesn't always look the way that you think it's going to look."

"我觉得安全就是这样一种东西 —— 嘴上谈它是一回事,真正把它做对完全是另一回事,因为它往往并不长成你以为的那个样子。"

"And it's just all about like always red teaming, always pentesting, always looking, always having a threat model, and then using that to figure out how is this thing going to get attacked, how are people going to get prompt injected."

"它的关键就在于:不停地红队演练(red teaming)、不停地渗透测试(pentesting)、不停地盯着、始终有一个威胁模型(threat model),再用它去推演 —— 这东西会怎么被攻击?人们会怎么被提示注入(prompt injection)?"

"Exactly."

"完全正确。"

"And I just feel like the team is just obsessed with this, and it's so important, because as a result I just trust the agent to run, and I can move on, and I can just have like a second agent. And if I didn't trust it, then I just wouldn't have been able to do that."

"我觉得整个团队都对这件事近乎执着,而这极其重要 —— 正因如此,我才敢放心让 agent 自己跑,然后转身去做别的、再开第二个 agent。要是我不信任它,这一切就根本无从谈起。"

"And internally, to actually get auto mode out to our users, we needed to really trust it first. And so what we did was we collected thousands of transcripts of like an entire agent trajectory and a permission prompt, and had auto mode classify whether or not it was safe. And it was extremely good at this."

"在内部,为了真正把 auto mode 交付给用户,我们得先彻底信任它。所以我们做的是:收集了数千份记录 —— 每一份都是一整段 agent 的行动轨迹外加一个权限确认 —— 让 auto mode 去判定它安全与否。结果它在这件事上表现极其出色。"

"So then we got red teamers, and we asked them to try to prompt inject and try to hack the codebase, and we used this to create evals, and made sure that all of these were denied."

"接着我们请来红队,让他们尝试做提示注入、尝试攻破代码库;我们用这些来构建评测集(evals),并确保这类攻击全部被拒。"

"And then we had our own internal teams try to prompt inject and hack Claude Code's auto mode, and then we improved auto mode to make sure that we caught all of these."

"然后我们又让自己内部的团队去尝试对 Claude Code 的 auto mode 做提示注入、做攻击,再据此不断改进 auto mode,确保把这些全都拦下来。"

"So it's not only just protecting you against the vulnerabilities that are out there in the wild today, but the most intelligent attacks that we can construct."

"所以它防的不只是当下野外已有的漏洞,还包括我们能构造出来的、最聪明的那一类攻击。"

"Yeah. I mean, it's honestly like a weird approach. I feel like there's all these features the last year where the first time someone pitched it, I was like, 'Ah, no, no way, that's not going to work.'"

"是啊。说真的,这是个挺怪的思路。过去一年里有好多功能,第一次有人提出来时,我都是:'啊,不行,绝对不可能,这没法成。'"

"And over time I just learned, like, I'm actually wrong like so often now, because building on the model is so weird."

"但慢慢地我学到一件事:我现在错的频率高得离谱,因为在模型之上做开发这件事,实在太反直觉了。"

"It's just like all this engineering stuff that I've learned over the years — like, so much of it I just have to throw out. And this is just like part of what the job is now. We're building on a new thing, and we just have to relearn it."

"我这些年学到的那一整套工程经验 —— 其中很大一部分,我现在不得不扔掉。而这恰恰成了这份工作的一部分:我们是在一个全新的东西之上做开发,只能重新把它学一遍。"

"And auto mode was definitely one of these. The first time I heard it, I was like, route the prompt to a model? No way, that's not going to work. And then it actually turns out empirically it works really, really well."

"Auto mode 绝对是其中之一。我第一次听到时心想:把 prompt 转交给一个模型?不可能,这没法成。结果从实践来看,它运作得非常非常好。"

"I heard you also love Loop."

"我听说你也很喜欢 Loop。"

"Um, yeah, I love Loop."

"嗯,对,我很喜欢 Loop。"

"How do you use it?"

"你怎么用它?"

"For Loop, there's this transition that we went through like a year and a half ago, where we were like, all right, there's source code, but actually the thing an engineer should interact with — maybe it's not the source code, maybe it's the agent."

"说到 Loop,我们大概一年半前经历过一次转变:当时我们意识到,有源代码没错,但工程师真正应该去打交道的对象 —— 也许根本不是源代码,而是 agent。"

"And so we made this leap of, like, I don't write the source code, I talk to an agent, and the agent writes the source code for me."

"于是我们完成了这么一跳:我不再亲自写源代码,而是跟一个 agent 对话,由它替我把源代码写出来。"

"And I think right now what's happening is we're making the next leap: I don't talk to an agent anymore, I talk to a loop, or I talk to a routine, and it prompts Claude for me. And it's just — it's crazy."

"而我觉得此刻正在发生的,是下一跳:我连 agent 都不直接对话了,我对一个 loop、或者一个 routine 说话,由它替我去给 Claude 下指令。这真的 —— 太疯狂了。"

"I mean, it's been like a year and a half, and this is like two big leaps."

"我是说,才过去一年半,这就已经是两次大跳跃了。"

"If you take like a step back, how are you seeing entire engineering orgs change?"

"如果往后退一步看,你观察到整个工程组织正在发生什么样的变化?"

"I'm going to put on my business cat hat. I have this like favorite case study — this is like a Harvard Business Review from the '90s, and they were talking about, like, computers are here, why are we not seeing the productivity benefits?"

"我要戴上我的'商业 cat 帽'了(谐音自己的名字 Cat)。我有个特别喜欢的案例研究 —— 是上世纪 90 年代的一篇《哈佛商业评论》,讲的是:电脑都来了,我们怎么还没看到生产力的提升?"

"And it's just this amazing snapshot into what it actually felt like at the time. People used to use mainframes; at some point companies switched to personal computers. It was sort of a new thing, and companies were trying to figure out how to use it — the same way they're trying to figure out how to use AI right now."

"它简直是当时那种心态的绝妙切片。人们原本用大型机,后来某个时点企业纷纷转向个人电脑。那是个新鲜玩意儿,企业都在摸索该怎么用 —— 就跟今天大家在摸索怎么用 AI 一模一样。"

"And it turned out that to get the productivity benefits from computers, what you had to do isn't, like, you have your paper filing cabinet and your paper-and-pen business process, and then there's a computer on the side that does something."

"结果发现,要从电脑里真正拿到生产力红利,你要做的并不是 —— 一边留着你的纸质档案柜、纸笔流程,一边在旁边摆一台电脑做点辅助。"

"Actually, what you have to do is you throw out the filing cabinet, you throw out all your paper and all your pens, and then you put a computer in the center, and everything has to run through the computer. It has to be at the center of every business process."

"真正要做的,是把档案柜扔掉,把所有纸和笔都扔掉,然后把电脑放到正中央,让所有事情都从电脑里走一遍。它必须处在每一条业务流程的中心。"

"And I feel like at Anthropic we do this thing where when you onboard, you don't ask people questions. No one asked me questions when they onboard — you probably have the same thing — they ask Claude. And this is kind of weird. This is the first company I've been at like that."

"而我感觉在 Anthropic,我们就有这么个做法:入职时,你不去问人。当初我入职没有去问别人 —— 你大概也一样 —— 大家都问 Claude。这有点怪,这是我待过的第一家这样的公司。"

"And I feel like for us, Claude is just at the center of everything. Whenever I have a question, I ask Claude. Whenever I write code, I use Claude. Whenever I need a code review, Claude does it. Whenever I need a security review, Claude does it. Whenever I need to fill out a form or something, Claude does it."

"对我们来说,Claude 就处在一切的中心。有问题,我问 Claude;写代码,我用 Claude;要 code review,Claude 来做;要安全审查,Claude 来做;要填个表单之类的,也是 Claude 来做。"

"So Claude is at the center of everything. And I feel like the companies that are really figuring it out — and there's a bunch of them now — they're just putting Claude at the center of it."

"所以 Claude 处在所有事情的中心。我觉得那些真正搞明白了的公司 —— 现在已经有一批了 —— 做的就是把 Claude 放到中心。"

"But I think for computers, the transition took 10 to 15 years. But actually for AI, because so much of our work is already digitized, and Claude can use a computer and it can write code and run code, this transition is happening a lot faster."

"不过对电脑来说,那场转变花了 10 到 15 年。但 AI 不一样 —— 因为我们的工作本就已经高度数字化,加上 Claude 既能操作计算机、又能写代码、跑代码,这场转变正在快得多地发生。"

"I think it's just really exciting. I feel like now I don't have to bug people anymore. And when I interact with people, it's because it's fun, and I get to collaborate with them on stuff, and we get to create something together."

"我觉得这真的特别让人兴奋。现在我不用再去打扰别人了。我跟人打交道,是因为那本身有趣 —— 我能和他们一起协作、一起创造点东西。"

"It's not that like I need something from them, because Claude can actually do a lot of that stuff now."

"不再是因为我有求于他们,因为很多那样的事,现在 Claude 其实就能做。"

"And I also feel like, as an engineer, I've just never had this much fun doing engineering, because the like tedious part I don't have to do. I'm just coming up with ideas, I'm talking to customers."

"而且作为一个工程师,我从没觉得搞工程能这么好玩过,因为那些枯燥的部分我不必再做了。我只管想点子、跟客户聊。"

"And every idea — I don't have a to-do list anymore. Claude just builds everything. And so my job is to come up with these ideas, and it's just so fun."

"每个想法 —— 我已经没有待办清单了。Claude 把一切都做出来。所以我的工作就是不断冒出这些点子,而这实在太有意思了。"

"Okay, so here's a question. Is the future product or engineering? Like, is everyone going to be a PM, or is everyone going to be an engineer?"

"那好,我抛个问题:未来是产品的天下,还是工程的天下?是人人都变成 PM,还是人人都变成工程师?"

"Everyone's going to be both. I feel pretty strongly that these roles are merging."

"两者都是。我相当确信:这些岗位正在融合。"

"When we look at our team, our product team all writes code, our devrel team all writes code, our design team all writes code. And then we look at our engineers, and a lot of them ship products end to end."

"看看我们团队:产品团队全员写代码,DevRel 团队全员写代码,设计团队全员写代码。再回头看我们的工程师,很多人是从头到尾把产品做出来交付。"

"They have an idea for what to build, they build it, they work with legal and marketing to figure out how we communicate this to the world and make sure it's safe, and with security too. And a lot of times they just see through this whole process end to end."

"他们有一个要做什么的想法,就把它做出来;还会和法务、市场一起搞清楚怎么对外传达、怎么确保它安全,也会拉上安全团队。很多时候,整个流程都是他们一手从头跟到尾。"

"So I think right now AI really benefits people who have a lot of curiosity, have a lot of product taste, who love to have this like end-to-end ownership. And now a lot of people are running like hundreds of agents. What are the products that you think people should be adopting as they transition from single to multiple to hundreds?"

"所以我觉得,当下 AI 真正受益的,是那些充满好奇心、有很强产品品味、又喜欢端到端把事情扛下来的人。而现在很多人手里同时跑着上百个 agent。在从'单个'到'多个'再到'上百个'的过渡里,你觉得人们应该用上哪些产品?"

"Until recently, the way that I wrote code was I had like six terminal tabs with six git checkouts of the same repo, and then I would just tab between them."

"直到不久前,我写代码的方式还是:开六个终端标签页,对同一个仓库 checkout 出六份,然后在它们之间来回切。"

"Now it's pretty different. I have like one tab. I use the new agent view that we just shipped. It's like so good, and I'm so glad that we took a while to iterate on it to make that really good."

"现在完全不同了。我就开一个标签页,用我们刚上线的新 agent view(智能体视图)。它太好用了,我很庆幸我们花了点时间反复打磨,把它做得真正好用。"

"And I also use the desktop app, because I don't have to fiddle with checkouts that way. It just does the work tree cloning — it creates the work trees for me."

"我还用桌面 app,因为这样就不用再折腾 checkout 了。它会自动处理 work tree 的克隆 —— 帮我把各个 work tree 都建好。"

"And the thing that I would not have expected six months ago is probably half my engineering now I do on my phone."

"而有件事我半年前绝对想不到:现在我大概有一半的工程活儿,是在手机上干的。"

"So I have so many agents running that I just start from my phone. I use remote control, which is amazing now. I'll start something on my computer, and then I'll just remote control in from my phone, and I'll just walk around — I'll get coffee, and then I'll check in on my agents, and maybe I'll start another agent."

"我同时跑着那么多 agent,干脆直接从手机上开。我用远程控制(remote control),它现在好用得惊人。我会在电脑上起一个任务,然后用手机远程连进去,接着就到处走 —— 去倒杯咖啡,顺手看看我那些 agent 的进展,可能再起一个新的 agent。"

"And sometimes I'm talking to someone and we come up with a new idea, I'll just start an agent on the spot. I'll talk to it with voice mode and just have it build something, and I don't even have to go back to my computer anymore."

"有时我正跟人聊着、突然冒出个新点子,就当场起一个 agent。我用语音模式跟它说话、让它直接做出来,我甚至都不用回到电脑前。"

"I remember when you started doing this, because you would actually leave work, have your computer on your desk open, plugged in, screen locked, and I just thought you would come back to the office at some point to get your computer."

"我记得你刚开始这么干的时候 —— 你居然会下班走人,电脑就开着摊在桌上、插着电、锁着屏。我还以为你迟早会回办公室来拿电脑。"

"But then it got pretty late, and I was like, hmm, maybe you just left it here by accident. And then it happened again the next day, and again the next day, and I was like, wait, this is so weird, because you're landing PRs, so your computer is right next to me."

"结果一直到挺晚,我就想:嗯,可能是不小心落这儿了吧。然后第二天又这样,第三天还这样,我就纳闷了:等等,这太奇怪了 —— 因为你在不停地合 PR,可你的电脑就摆在我旁边没动。"

"And I remember you responding, and you're like, 'Yeah, I'm coding from my couch.'"

"我还记得你回我的话:'对啊,我在沙发上写代码呢。'"

"Yeah, that was the week that remote control got really good."

"对,那正好是远程控制变得特别好用的那一周。"

"Yeah. So another thing that users are asking about all the time is: how do you do context engineering, especially in a large enterprise?"

"嗯。还有一件用户老是问的事:你们是怎么做上下文工程(context engineering)的,尤其是在大型企业里?"

"This is a thing — people used to talk about prompt engineering, they used to talk about context engineering. This was sort of matching where the model was at the time."

"这事是这样的 —— 人们以前谈提示工程(prompt engineering),后来谈上下文工程。这其实是在跟当时模型所处的阶段相匹配。"

"Back in the days of Sonnet 3.5, you had to prompt engineer. Back in the days of Opus 4, you had to context engineer. But with the models of today, you don't do any of this."

"在 Sonnet 3.5 那会儿,你得做提示工程;在 Opus 4 那会儿,你得做上下文工程。但放到今天的模型上,这些你都不用做了。"

"You give it the minimal possible system prompt, the minimal possible tools, and then you let the model figure it out. Like, you just have to give the model some way to pull in the context. I think that's the most important thing. How do you think about it?"

"你给它尽可能精简的系统提示、尽可能精简的工具,然后让模型自己去搞定。你只需要给模型一个能把上下文拉进来的途径就行。我觉得这才是最重要的。你怎么看?"

"I see things very similarly. I'm a context minimalist. So my general philosophy is, tell the model only what it needs to know and let it figure out the rest of it."

"我的看法非常相似。我是个'上下文极简主义者'。我的总原则是:只告诉模型它需要知道的,剩下的让它自己去摸索。"

"I think when you give the model too much context, it's kind of like you're micromanaging it. And sometimes the model knows a better way to get to the same outcome, and I personally prefer to give the model that freedom to do that."

"我觉得,当你给模型塞太多上下文,就有点像在微观管理它。而有时模型其实知道一条通往同样结果的更好路径,我个人更愿意把这份自由留给它。"

"And then in general, we're also making our harness more lean, so that you have more room for your own prompts, and so that it follows your prompt better."

"另外整体上,我们也在把我们的 harness(运行框架)做得更精简,好让你有更多空间放自己的 prompt,也让它能更好地听你的指令。"

"There's all these different ways to use Claude now. But I feel like in a year it's going to be a totally new set of things, and it's going to be so surprising if it's still these same things, because I think we're seeing these giant trends happening right now."

"现在用 Claude 的方式五花八门。但我觉得一年后会是一整套全新的玩法 —— 要是那时还是这些老一套,反倒会让人意外,因为我看到此刻正有几股巨大的趋势在发生。"

"Agents are running for longer. They're more autonomous. Very rarely am I running one agent at a time — it's usually like a few agents, or dozens, or hundreds, or thousands."

"Agent 运行的时间越来越长,也越来越自主。我现在很少只跑一个 agent —— 通常是好几个,或者几十个、几百个、上千个。"

"And so the form factor for that is going to be really different than what came before. And I don't know what it's going to be. And I think in large part it's going to be up to the team to figure it out."

"所以承载这种用法的形态(form factor),会和以前的大不一样。它会变成什么样,我并不知道。而我觉得,这在很大程度上要靠团队自己去摸索。"

"And this is why I'm so happy we run the team the way that we do, where everyone just comes up with ideas, and everyone is able to think about the product. Everyone talks to users all the time, because I don't think these ideas are going to come from us — it's going to come from the team."

"这正是我特别庆幸我们这样带团队的原因:人人都在冒点子,人人都能去思考产品,人人都在不停地和用户聊。因为我觉得这些想法不会来自我们这几个人 —— 它会从整个团队里长出来。"

"Totally. And from everyone in our community building with us."

"完全同意。也会从我们社区里、所有和我们一起在做东西的人那里长出来。"