本集简介
双语字幕
仅展示文本字幕，不包含中文音频；想边听边看，请使用 Bayt 播客 App。
你正在学习如何进行复合工程。
You're figuring out how to do compounding engineering.
团队里只有两个人，但感觉就像有十五个人一样。
There's two people on the team, but it really feels like there's 15.
用AI编程不仅仅是编码部分。
Coding with AI is more than just a coding part.
将其用于研究、工作流程，应该用在所有方面。
Utilizing it for research, for workflows, It should be used for everything.
我们过去三周没有使用过WinServe或Cursor。
We haven't touched WinServe or Cursor in the last three weeks.
这两者都具备代理式编码能力，但Plot更进一步，将复杂度降低了十倍。
Both of those have agentic coding capabilities, but Plot just takes it one step further by simplifying it by a factor of 10.
我们一直在大力推动让AI替我们工作，我们只是在管理AI。
We've been really leaning into let's AI do the work for us, we're just managing the AI.
你首先花时间构建了一个能有效生成其他提示的提示。
What you did first is spent time building a prompt that effectively builds other prompts.
这太酷了。
That's really cool.
基兰、尼特什，欢迎来到节目。
Kieran, Nitesh, welcome to the show.
谢谢。
Thank you.
谢谢你邀请我们，汤姆。
So much for having us, Tom.
很高兴有你们加入。
I'm psyched to have you.
对于不了解的人，你们两位都在Quora工作，那是Every的AI邮件助手。
So for people who don't know, both of you work on Quora, which is Every's AI email assistant.
基兰，你是总经理。
Kieran, you're the GM.
尼特什，你是工程师。
Nitesh, you're an engineer.
除了Cora是一个非常棒的产品，而且我非常兴奋能将它带给所有收听或观看这个节目的人之外，我之所以想邀请你们两位来做一期节目，是因为我认为你们正在探索一种全新的工程方式。
And beyond the fact that Cora is a really cool product, and I'm really excited to bring that to everybody who listens to this show or watches this show, I wanted to do an episode with the two of you because I think that you're figuring out a new way to do engineering.
因为实际上，Cora团队只有两个人，但它给人的感觉却像是有十五个人在工作——有代理自动拉取代码评审、处理分支，然后你们再推送上去，其他代理负责审查，这一切简直太疯狂了，这是一种全新的软件构建方式。
Because really, Cora has, you know, there's two people on the team, but it really feels like there's 15 because you've got agents who are pulling down PRs and working on branches, then you're pushing them up and other agents reviewing, and it's just this kind of crazy thing that it's a new way to build software.
Kieran，你前几天说过一句话让我印象深刻，你说你们正在探索如何实现‘复利式工程’，也就是说，每完成一项工作，你们都在让下一项工作变得更轻松。
Kieran, you said something the other day that really stuck with me, which is you're figuring out how to do compounding engineering, so with each piece of work you do, you're making it easier to do the next piece of work.
我觉得把你们所学到的经验分享给所有观看节目的人非常重要，因为我们有了新工具，因此也需要新的原则和新的工作流程来使用这些工具。
And I just think that it's really important to bring what you guys are learning to everybody that watches the show because it's like we we have new tools, and so we need new principles and new workflows for using those tools.
所以我非常期待和你们聊聊这个话题。
And so I'm really excited to talk to you about that.
是的，谢谢。
Yeah, thanks.
构建Quora真的非常有趣，但更重要的是，成为Every的一员，能够接触到这些工具、思维模式，以及令人兴奋的新工作方式，真的帮助我们重新思考如何构建产品。
It's really fun to build Quora, but like being part of every and like being in an environment where you get access to tools, like access to thinking, access to exciting new ways to work really helps us rethink how we build.
所以这其实是一次实验。
So like, it's really an experiment.
我们在开发一个产品，Quora，但同时也在探索应该如何构建它。
We're building a product, Quora, but at the same time we're figuring out how we should build.
这非常有趣。
And that's super interesting.
我们正处在这样一个阶段：人们会问，你对这个新模型怎么看？
And we're like right in the middle where people say, what do you think of this new model?
比如，我们该如何使用这个研究工具？
Like like, how do we use this research tool?
我们只是在不断尝试各种方法。
And we're just trying things out.
过去几周，我和娜塔莎明显感受到一种转变，事情正在变化，而且我们不是唯一这样觉得的人。
And Natasha and I, we've been like really feeling a shift in the last weeks, I would say, where we're like, like things are changing and we're not the only ones.
我们也能听到其他人这么说，但这样的人还不多。
Like we hear other people say that as well, but not a lot of people.
我们学到了很多，想分享一些我们学到的经验。
And what we've learned is a lot and we want to share a little bit of what we learned.
而且我们意识到，我们才刚刚开始。
And also what we know is like we're just barely starting.
我们只是触及了这件事的表面。
We're scratching the surface of this.
这是一个正在发生的巨大转变，由新模型驱动，由人们的思维方式驱动，由MCP驱动，总之变化非常多。
And it's a big shift that's happening right now by new models, by how people think, by MCP, by like, just it's a lot.
而且，是的，从不同角度来讨论这个问题真的很好。
And, yeah, like, it's great to talk about that from different perspectives.
是的。
Yeah.
对。
Yeah.
我同意。
I agree.
而且我觉得在Every这里真的很特别，因为我们几乎每天都会在Discord上看到新人说，我构建了一个AI智能体。
And and I think it is so special to be at Every because we do like, every day, there's someone new in the Discord who's like, I built an AI agent.
你想要在我们上线前先用一下吗？
Do you wanna, like, use it before we launch it?
所以我们能提前接触到OpenAI的模型，有时甚至能提前使用Anthropic的模型。
And so, you know, we get access to OpenAI models before they come out and and sometimes anthropic models.
所以我们有这种早期优势，而你们又特别擅长如何将这些模型真正融入生产流程中。
And so we have this, like, early edge, and then and then you guys are so good at figuring out how to actually incorporate them into, like, a production process.
你之前提到Kieran，说有些事情发生了变化。
So you said Kieran that something changed.
所以我想了解一下，你觉得哪些方面发生了变化，以及你们的流程大致开始朝什么方向发展。
So I guess I I wanna get a sense of what you think changed and what draw the broad strokes of what the workflow is that's starting to emerge for you guys.
是的。
Yeah.
对我来说，显然是所有事情都融合在了一起，但我觉得最重要的是我意识到，用AI编程不仅仅是写代码的部分。
For me, like, obviously it's like everything coming together, but I think the biggest thing is a realization in myself that coding with AI is more than just a coding part.
它真正关乎的是将其用于研究、流程以及方方面面。
And it's really about utilizing it for research, for workflows, for everything.
它应该被用于一切。
Like it should be used for everything.
我们现在到了一个阶段，智能体已经足够强大，可以真正完成所有任务。
And we're now at a point where the agents are good enough that they can actually do everything.
所以我们需要重新思考，比如 Cursor、Windsurf，传统的编码方式固然很好。
So we need to rethink again, like, hey, cursor, windsurf, like the old school way of coding was great.
更像是那种随性编码的风格。
Like more of the vibe coding.
那是一步进步。
Like that was one step.
而现在我们意识到，实际上我们只需要给出一个任务，它就能完成。
And then now it's the realization, oh, actually we can just give a task and it will do it.
但工作仍然需要有人来做，比如我们该做什么？
But still the work needs to be done by like, what do we do?
我们该如何做？
How do we do it?
而且我们意识到，应该更深入地拥抱这一点。
And just a realization that we should lean into that more and really go deep.
这就像是披着一件布外套。
And and it's like cloth coat.
就像现在有了优秀的编码代理，它们能很好地配合新模型，比如在遵循指令方面表现得非常出色。
Like it's just good coding agents or agents available that actually start to work with new models like cloth for like really good at following directions and instructions.
所有这些因素正在汇聚在一起。
And it's all of that coming together.
我意识到，我们已经到达这里了。
That's that I realized like, oh, we're here.
未来已经到来。
Like the future is here.
我们一直谈论的这种代理式演进。
This thing we've been talking about that was going to be the agentic evolution.
突然间它真的实现了，并且在现实世界、非实验性的场景中发挥作用了。
Suddenly it works and it's working in real world, non experimental playing thing.
这就像是我们在开发一个应用，而这个应用正在顺利构建中。
It's just like we're building an app, and it's working building the app.
所以我的理解是，这不仅仅是关于使用AI进行开发。
So what I'm hearing is like, it's not just about developing with AI.
而是所有那些你用AI来辅助的开发工作，而你用得最多的工具就是Cloud Code。
It's all the things that go into developing that you're using AI for, and that the the thing that you're using the most for this is Cloud Code.
对吗？
Is that right?
如果没错的话，请给我介绍一下，对于那些不了解或没用过Cloud Code的人，先简单讲讲什么是Cloud Code，然后再具体说说你是怎么用它的。
And if it's right, like, tell tell me about, like for people who don't know Cloud Code or haven't used it, give us a little introduction to Cloud Code and then tell us about, like, exactly how you're using it.
是的。
Yeah.
Cloud Code本质上是Anthropic推出的编码代理工具，底层基于Cloud技术。
Cloud Code is basically the coding agent version from Anthropic that uses Cloud under the hood.
它以命令行工具的形式运行在你的终端里，这有点像
And it runs in your in your terminal, as a CLI tool, which is kind of a
你愿意共享屏幕给我们看看吗？
Do you wanna share your screen and like show us?
好的。
Yeah.
好的。
Yeah.
Cloud Code 是一个你在终端中使用的工具。
So cloth code is a tool that you use in your terminal.
我知道对于非技术人员来说，这听起来有点吓人。
And I know for nontechnical people, this is like, oh, this is scary.
但我已经成功让一些非技术背景的朋友使用 Cloud Code，他们都觉得这太棒了。
But I've converted friends who were not technical to use Cloth Code and they were like, oh, this is great.
但它其实非常简单。
But it's really simple.
你只需打开终端，输入 Claude，就会弹出一个界面。
You just hit, you start your terminal, you say Claude, and an interface will pop up.
对于那些在听而不是看的人，他现在正在使用终端。
And basically, for people who are who are listening instead of watching, he's in his terminal.
那是一个经典的黑色屏幕，看起来就像你在用DOS系统一样，他只是输入了‘Claude’。
It's, you know, the the classic black screen that you're, you know, feels like you're using DOS or something, and he just typed Claude.
然后我们看到一个提示，写着‘欢迎使用Claude Code’，并出现了一个小文本框，供他输入任何命令。
And then we just got a a thing that says welcome to Claude code, and there's a little text box for him to type in any command.
是的。
Yeah.
这有什么不同？是什么让它与众不同？
And why is this different or what makes this different?
它可以访问目录或计算机。
This has access to the directory or the computer.
因此，它能查看我电脑上已有的文件。
So it can look through files on my computer already.
它还能在我的电脑上运行程序。
It can run things on my computer.
它可以截取网站的截图。
It can take screenshots of websites.
它可以搜索网页。
It can search the web.
它拥有各种工具，而且比普通版的Clot工具多得多。
Like it has tools, but way more tools than available in a normal clot version.
这一点很重要，因为工程工作——比如构建东西——确实需要比基础功能更多的工具。
And that's important because engineering work, like building stuff, you do need more tools than just like the basics.
你需要GitHub来看你需要构建什么、当前状态如何，或者CI流水线是否通过了测试。
You need GitHub to see what you need to build or what the status is or what the CI pipeline does, like do the test fail.
所有这些功能都集成在一个编码代理中，这让我实际的工作流程或日常操作能够由代理完成。
Like having all these things available in one coding agent actually makes it possible for me to have a workflow or like a thing I do actually be done by an agent.
这才是关键所在。
And that's that's the important thing.
真正重要的是，这个代理所做的远不止编码，因为如果你问工程师，大部分工作可能确实是编码，但也许只占20%，而80%的工作其实是弄清楚下一步该做什么，或者理解他人的反馈并加以解读。
Like really the compound word comes in by doing more than just coding because lots of like if you talk to an engineer, like most of the work is maybe coding, but maybe it's actually 20%, maybe 80% of the work is like figuring out what to do next or understanding what people like, what their feedback is and how to interpret it.
你可以在这里做的是，比如一个有趣的方式是，你说：我们上周发布了什么？
And what you can do here is you can, for example, like a fun way is like to use it, you say, let's say, what did we ship in the last week?
所以它知道这些信息。
So like it knows stuff.
所以我问它我们发布了什么，它很可能会查看 Git 日志，因为这是我们追踪已发布内容的方式。
So I'm asking it what we shipped and it will most likely look at the Git log because that's how we track what we did ship.
是的，所以它会查看 Git 日志。
And yeah, so it looks through the Git log.
它会查看我们合并到主分支的内容。
It looks at what we merged to main.
是的，这是一种使用它的有趣方式。
And yeah, that's a fun way to use it.
比如，我们可以用它来做产品营销，它会说：这些是错误修复、简要跳过功能、聊天面板状态、邮件摘要、XML 标签、主要功能、简要健康监控、时区自动检测。
And for example, we can use this for product marketing and it says, these are the bug fixes, a brief skip functionality, chat panel state, email summary, XML tags, major features, brief health monitoring, time zone, auto detection.
这些都是我们发布的内容。
These are all things we released.
现在我可以这么说
And now I can say
而且它被写成了一份很好的总结，是的。
And it's written in in in a nice write up that Yeah.
这很技术性。
It's technical.
是的。
Yeah.
但对两个人来说，这其实太多了。
And it's actually a lot for for two people.
有什么？
There's there's what?
比如，六个主要功能，五个重要修复，还有三个基础设施更新。
Like, six major features and, like, five important bug fixes and three infrastructure updates.
这太多了。
Like, that's a lot.
是的。
Yes.
这确实很多。
It it's a lot.
而且像这一周，我们一直在大力推动让AI替我们工作，我们只是管理AI。
And and like this week, we've been really leaning into like, let's AI do the work for us and we're just managing the AI.
另一件事，比如，如果有人来找你，说：‘这个的进展如何？’
One other thing, for example, is if if you have someone come to you like, oh, what is the status on this?
或者：‘你们下周要发布什么？'
Or like, what are you going to ship next week?
我们来看看AI会怎么做。
Let's see what it will do.
你能看到当前的待办列表和即将发布的内容吗？
Can you see what is in the pipeline and what will come out soon?
这太棒了。
So this is awesome.
另外，Nitesh，在这个过程中，如果你有任何想法，随时可以加入。
And, Nitesh, while this is going, if you if you wanna jump in at any point, feel free to.
我会在某个时候把话题抛给你，但你也别客气，随时都可以插话。
At some point, I'll I'll lob it to you, but also, like, just I'll you know, feel free to jump in.
好的。
Yeah.
当然。
Sure.
对。
Yeah.
那我们拭目以待。
So we'll see.
我不确定它有没有项目访问权限，但你明白我的意思。
Like, I don't know if it has project access, but like, you you get the gist.
如果你把信息和智能体连接起来，使用起来就非常简单。
Like, if you have the information connected to the agent, it's very easy to use it.
使用你熟悉的工具非常重要。
And it's very important to use a tool you're familiar with.
到目前为止，我认为Clothecode最适合我。
And at this point, I think Clothecode works the best for me.
它最具灵活性，因为它不仅仅解决编码问题。
It is the most flexible because it doesn't only solve coding issues.
这一点很重要。
And that's important.
许多编码代理只是为了编码而设计，但我想做的不止是编码。
Lots of these coding agents are made to code, but I want to do more than coding.
我希望它能成为工程方面的通用支持工具。
I want it to be like a support in engineering in general.
我认为Clot团队确实考虑到了这一点。
And I think the Clot team really thought about that.
他们没有把它做得太具体，而是保持了通用性，同时在解决问题、观察其行为、思考错误并自我修正方面表现得非常出色。
They made it not too specific and they kept it general while actually being really good at solving things and looking at what it did, thinking about the mistakes it made and self correcting.
所以这些都在逐渐整合在一起。
So that is stuff coming together.
这非常困难。
That's very hard.
这让它现在可以被使用了。
That makes it possible to, use now.
是的。
Yeah.
编码、Cursor 和智能编码之间有什么区别？
What's the difference between coding and cursor and agentic coding?
Claude Code 与我们熟悉的 Cursor 和 Vinsurf 相比，是一个非常简单的突破。
Claude code is such a simple departure from the Cursor and Vinsurf that we're used to.
这两者都具备智能编码功能，但 Claude 只是进一步简化了它，我认为简化了大约十倍。
Like, both of those have agentic coding capabilities, but Claude just takes it one step further by simplifying it by, I I think, like, a factor of 10.
所以，基兰之前提到的，Clord Code 可能会让人感到畏惧，因为它是一个终端。
So, what Kieran was, telling earlier about how, Clord code may feel intimidating because it is a terminal.
但事实上，它比WinSurf和Cursor简单得多，因为这里除了文本按钮和文本框之外什么都没有。
But in reality, it is, like, so much simpler than the WinSurf and cursor because there is nothing except text button text box here.
对吧？
Right?
没有任何命令键、快捷键、接受、删除、拒绝或移除按钮。
There's no no, like, command key, no shortcuts, no accept, delete, reject, remove.
真的什么都没有。
Like, there's nothing.
它就是一个文本框，之所以能工作，是因为底层的Claude模型现在强大得多。
It's just a a text box, and it works because the model the underlying Claude model is so much more capable now.
所以它能够持续更长时间并执行工具调用。
So it's able to work for longer and do tool calls.
所以这是一个更简单的界面，同时却更强大。
So it it's it's like a simpler UI, which makes it at the same time more powerful.
尽管像Cursor和Clotco背后的底层模型其实是一样的。
Even though even though like the underlying model behind like Cursor and Clotco is the same.
是的。
Yeah.
一个例子就是今天早上，我在提取一些指标数据。
And an example of this is this morning, I was pooling some metrics.
我当时想，为什么我们没有收到任何
I was like, why didn't we get any
这个表单的回复？
responses to this form?
然后，为了说明一下背景，我们有一个表单，用来询问用户如果无法再使用Chorus会有多失望，以此判断我们的表现如何。
And then And for for for context, like, basically, we have a form that we ask people how disappointed they would be if they could no longer use Chorus, we can tell how how well we're doing.
你注意到我们每周都会开一次会，回顾所有指标，而且你也发现没有人填写过这个表单。
And you noticed we have a we have a weekly meeting where we go through all the metrics, and we know that you noticed that no one had filled out that form.
所以你就进入Cloud Code，问：为什么没人填写这个表单？
So you're going into Cloud Code, and you're asking, hey, like, why is no one filling out this form?
是的。
Yeah.
我当时就想，肯定有什么问题，这个表单根本没发出去。
I was like, there has to be something like this form was not sent.
于是我问了句：十四天前，是不是出了什么问题？
And I asked like, hey, fourteen days ago, something went wrong.
你能看看当时出了什么状况吗？
Can you see what went?
它做了一个待办清单，比如获取控制器的最新日志变更、搜索代码库。
And what it did, it made a checklist to dos like fetching recent log changes to the controller, searching the code base.
所以它查看了那个日期前后有哪些改动，发现我们移除了一段用于添加用户的代码，这里写着：嘿，其实你只需要加上这个。
So it looked through what changed around that date and it found we removed a piece of codes that adds people there, which is here, it says, hey, actually you just need to add this.
我说：好吧，帮我执行一下。
And I said, okay, do it for me.
创建一个拉取请求。
Create a pull request.
它真的做到了。
And it did that.
我说，哦对了，顺便说一下，我还要写个脚本，把我们漏掉的人都加进去，完成迁移。
And I said, oh yeah, by the way, I'm also going to create a script that will then add everyone that we missed to it, migrate it.
就这样结束了。
And that was it.
有趣的是，我根本没花什么精力。
And the fun part was like, I didn't it didn't cost me any energy.
就像我在GitHub上记下来，以后再看一样简单。
Like, it was as easy as me writing it down in GitHub to look at later.
我不需要。
I don't need to.
我只要问一下，它就会立刻完成，这真的很棒。
I just ask it and it does it immediately, which is really nice.
就像收件箱清零一样，耗时不到五分钟？
It's like the inbox zero, does it take less than five minutes?
就这种类型的事情。
Do it kind of thing.
是的。
Yeah.
我认为人们可能没有完全意识到，如果没有AI，这个任务可能需要花费从三十分钟到几个小时不等的时间。
I think the thing that people may not fully realize is that that's a thing that task could take anywhere from, like, thirty minutes to a couple hours without without AI.
而且还不只是这样。
And it's not just that.
它需要你集中注意力，专门抽出时间坐下来去做。
It would require you to, like, focus on it and, like, put aside time to sit down and do it.
而现在，你只需发送这样的请求，然后就可以接着发送另一个、再一个。
And now you just sort of send off requests like that, and then you can send off another one and another one.
你可以同时运行多个这样的任务。
You have a bunch of these working in parallel.
所以给我具体描述一下，你的实际工作流程是怎样的？你到底在做什么？开了多少个标签页？你真的在手动写代码吗？是不是同时运行着五个任务？你只是在用Cloud Code吗？给我讲讲具体情况。
So give me a snapshot of what that looks like concretely, what your actual workflow is, what are you actually doing, how many tabs do you have open, like, are you actually doing any hand coding yourself, do you have like five in parallel, are you just using Cloud Code, like, give give us a give me a sense of that.
对。
Yeah.
我也会给你看看我的屏幕。
I'll show you my screen as well.
也许，尼特什，你可以告诉我们之前我们刚获得云服务早期访问权限时做了什么，当时我们有多兴奋。
Maybe, Nitesh, you can you can tell what we did before, like, when we got early access to Cloud, like, we were excited to what we did.
我来共享一下我的屏幕。
I'll show share my screen.
是的。
Yeah.
对。
Yeah.
所以，这是在直播触发前一天的情况。
So, this is like, like one day before, the live stream was triggered.
我们当时想，好吧，明天编程就要变了。
We were like, okay, tomorrow, coding is gonna change.
我们会拥有一个强大得多的模型，能够胜任我们所有需要的工作。
We'll have a much more capable model, which will be able to work for everything that we want.
我们基本上会得到一个为我们服务的编程精灵。
And we're basically gonna get like a coding genie for us.
所以，今天我们最高效的做法，不是进行常规的编程任务，而是进行两小时的头脑风暴，列出一份庞大的问题清单，让明天更强大的模型来解决。
So the best, most productive thing for us to do today, instead of doing our regular scheduled programming, we should just jam for a two hour call where we make a massive list of issues that we want the future, like tomorrow's superior model to solve.
我们确实这么做了。
And we did that.
我们创建了大约20个问题，涉及我们想修复的内容、原本计划要做的工作，并为新的云模型做好了准备。
We we created like 20 issues in terms of, you know, like what we want to fix, what were the things that we were planning to work on, and prepared the system for the new cloud model.
是的。
Yeah.
有趣的是，Nitesh 让 ChatGPT 说：‘嘿，明天我们已经实现了通用人工智能。’
And it was funny because Nitesh had like, he prompted ChatGPT to say, hey, tomorrow we have a we reached AGI.
你能帮我们列出所有需要做的事情，准备好让通用人工智能解决我们所有的问题吗？
Can you can you, yeah, can you help us come up with everything we need to do and like prepare the AGI to solve everything we did.
然后我们把这内容输入到 Anthropic 的提示词优化器，用它作为提示词，接着我们创建了——等等。
And then we fed that into the prompt improver of Anthropic, and then we used that as a prompt and we created Wait.
在你继续之前。
Before you move on.
在你继续之前。
Before you move on.
对于正在收听的朋友们来说，基本上你们在 GitHub 里有一个类似 Trello 的看板，针对每个你们确定要做的事情，你们都准备了一份详细文档，说明这是个功能、修复漏洞，或者其他类型的任务，以及具体如何执行。
So so for people for people who are listening, so basically, you have this sort of Trello board type thing inside of GitHub, Kanban board, and for each thing that you've identified as what you wanna do, it looks like you have a document that lays out in detail, okay, if it's a feature or it's like a bug fix or whatever, lays out in detail what it is and how to actually do it.
你能打开其中一个看看吗？
Can you open up one one of them?
好的。
Okay.
比如一个功能是让 AI 生成合成数据，这份文档包含了从问题陈述、解决方案愿景，到所有需求和技术细节的完整内容。
So like a feature is you wanna generate you wanna have AI generate synthetic data, And it has this document has everything from a problem statement to, like, a solution vision to all the requirements and all the technical requirements and, like, a bunch of a bunch of stuff.
而且它甚至还有实施步骤，包括预计耗时天数之类的。
But it's and even it has seems like it has implementation steps with day counts and stuff like that.
这挺有意思的。
So Which is funny.
这是一个一
This is a One
一天就像一秒钟。
day is like one second.
好的。
Okay.
是的。
Yeah.
是的。
Yeah.
是的。
Yeah.
所以我们使用封闭代码，我们创建了这些自定义提示来生成这些内容，因为创建这些东西非常耗时，即使使用ChatGPT，也需要完成很多步骤，要查看所有代码，需要去
So this we use closed code, we have this custom prompts that we generated to create these because like it's a lot of work to create these and even with ChatGPT, like there's a lot of steps you need to look at all the codes, need to go
仔细思考。
to have think about it.
是的，这需要大量思考，所以很难做好。
Yeah, there's a lot of thinking, so it's really hard to do it well.
所以我们做了个云代码命令，命令是一种你经常使用的自定义提示。
So what we did, we created a command in Cloud Code, a command is kind of a custom prompt that you use a lot.
我们的命令是：嘿，有个功能。
And ours is like, hey, there's a feature.
抱歉，这个命令是在云代码里，还是在Cursor里？
Sorry, is a command in Cloud Code or a command in Cursor?
因为你打开了Cursor。
Because you have Cursor open.
是的。
Yeah.
我打开Cursor是因为我要用它编辑文件，但这是Claude代码。
I have cursor open because that's how I edit files, but it's Claude code.
所以你看，我们现在在Claude里，可以通过按CCY来使用这个命令，也就是Claude代码。
So you can see we're in Claude, and and I can use this command by hitting CCY, which is Claude code.
然后我会说一些我遇到的问题，比如一个bug、一个问题，或者任何事情。
And then I say something like a problem I have, like a bug, a problem, anything.
所以这个过程非常顺畅。
So it's very low friction.
所以我有这个CCI命令，我和娜塔莎刚刚一起快速协作。
So I have this CCI command and Natasha and I were just jamming.
我们说：‘如果我们这样做会怎么样？’
We're like, oh, what if we do this?
哦，这听起来很棒。
Oh, that sounds cool.
然后通过语音转文字，它就开始运行了。
And then a voice to text and it starts.
那我们来看看这个是怎么运作的。
So so let's let's see how this works.
在它运行的时候，我们可以一起回顾这个过程。
And then while it's running, we can go over the thing.
我希望在知乎上实现无限滚动，当我浏览到一个简短回答的末尾时，它应该自动加载下一个回答，直到所有未读的回答都被读完。
So I want infinite scroll in Quora where if I am at the end of a brief, it should load the next brief, and it should go until every brief that's unread is read.
所以，就像这样，
So, like So,
是的，我只是希望人们明白，基兰几乎从不打字，全部使用语音转文字。
yeah, I I just want people to understand, like, Kieran almost never types anything and and does all voice to text.
所以他当时正在用语音转文字输入到他的终端，通过Cloud Code，使用一个名为Monologue的内部工具——据我所知，这是一个尚未发布的内部项目，而她是该工具的第四大用户，但目前仍处于保密状态，不过这里会稍后透露一点预览。
So he was just doing voice to text into his his terminal, into Cloud Code with, I believe, an internal, as of yet unreleased internal every incubation called Monologue, which she is the number four biggest user of, but still still under wraps, but but, you know, a little preview in here coming soon.
基本上，它看起来是在将这些内容转换成我们之前看到的那份文档，还是直接去执行它呢？
And, basically, what what it seems like it's doing is it's it's taking that is it turning is it turning that into the that document that we were looking at earlier or is it actually going and executing it?
是的。
Yeah.
它会把我刚才说的话插入到未来的描述中，然后依次执行这些步骤。
So what it does is it will insert whatever I said here in the future description, and then it will follow all these steps.
这些步骤包括研究和最佳实践。
And these steps are research, research best practices.
首先是基于代码库进行定位。
So one is grounding itself in the code base.
所以先研究现有内容，然后研究最佳实践。
So researching what exists, then it's researching best practices.
它会搜索网络和开源模式。
So it's searching the web, open source patterns.
所以它是在一般性的最佳实践中进行定位。
So it's like grounding it in like best practices in general.
然后它会提出一个计划。
Then it's will present a plan.
当我 saying '嗯，听起来不错'的时候。
And when I say, yep, sounds good.
我喜欢在计划阶段保留人为审核，因为有时它会出错，但大多数时候是对的。
Like, I like that review human in the loop for the plan because sometimes it does it wrong, but most of the time it's right.
然后我会说，'嗯，听起来不错'。
Then I say, yep, sounds good.
然后它会创建 GitHub 问题，并将其放入正确的栏目中等等。
And then it creates the GitHub issue and it will put it in the right lane and all that.
哦，有意思。
Oh, interesting.
所以就像我们之前在 GitHub 上看到的整个看板系统，你通过语音将功能输入到 Cloud Code，然后它会进行所有研究，生成那份长文档，并直接添加到 GitHub 问题中。
So it's like that whole Kanban we were looking at in GitHub earlier, you've created a way for you you speak your feature into Cloud Code, and then it does all the research to create that long document, and then just adds it into into GitHub issues.
这太酷了。
That's really cool.
是的。
Yeah.
这是一个重要的步骤，因为这和 Cursor 编码不同，在 Cursor 中，你通常会跳过这一步，因为那个工具并不是为此设计的。
It's it's an important step because it's like, this is different from cursor coding because in cursor, normally you skip this step because the tool is not really made for it.
这个工具主要是用来写代码的。
Like, the tool's made to code.
是的，你可以创建 Markdown 文件之类的。
Yes, you can do create you can create markdown files and all of that.
但让我们聚焦于问题跟踪器吧。
But let's lean into like an issue tracker.
它已经存在，运行良好，人们都在使用它，而且它已经整合了现有的工作流程。
It exists and it works well and people use it and it already hooks into existing like, patterns.
我们可以把这个交给开发者，他们就能直接实现。
Like, it like, we we can give this to a developer and they can implement it.
是的。
Yeah.
有一点我要指出的是，你正在运行这个系统，我记得我们第一次看到Opus 4时，简直惊呆了，因为它能持续运行而无需任何人工干预，最终给出相当不错的结果。我们之前也做过一些类似智能体的东西，但它的自主性和质量完全不是一个层级。
And one of the things that just to point out is, like, you're running this, and I I think one of the special things that when we saw Opus four for the first time, we were like, holy shit, is that it just runs forever with without any intervention and then gives you a pretty good result, which we've had sort of agentic type things for a little while, but it it's just a way different level of autonomy and quality than than we've ever had before.
它就像是在逐一完成这个待办清单，而其他智能体循环在细致程度上恐怕都远远不及。
And and it's like just checking things off of this to do list in a way that I think other agent loops are just gonna be a lot less thorough.
对。
Yes.
当然。
Absolutely.
我和基兰之间有个有趣的小比赛。
Me and Kieran have like a fun thing going on there.
我们正在比谁能让ClorCode运行更久。
We're trying to see who can have ClorCode running for the maximum amount of time.
基兰现在停止了运行。
Kieran is stopping the list right now.
他跑了二十五分钟。
He Twenty five minutes.
他运行了二十五分钟。
He ran it for twenty five minutes.
我现在才跑了八分钟。
I'm only at eight minutes right now.
天啊。
Oh, man.
基兰，你是怎么让它运行这么久的？
How, how did you get it to go so long, Kieran?
这是一个非常、非常长的计划，包括了嗯。
A very, very long plan, includes yeah.
这只是一个非常、非常复杂的长期计划，还包含大量测试，确保运行所有测试并修复所有问题。
It's just very, very complicated long plan and also include a lot of tests and just make sure that it runs all the tests and fixes all the tests.
有意思。
Interesting.
它持续的时间相当长。
It goes pretty long.
是的。
Yeah.
嗯。
Mhmm.
让我听听，等等。
Take me through wait.
我想了解一下。
I wanna understand.
你是怎么做出那个生成提示的提示的？
How did you make that prompt that that creates the prompt?
或者说是生成研究文档的那个提示？
Or, like, the prompt that creates the research document?
那么，你是怎么知道该包含哪些元素的？
So, like, how did you know which elements to put in?
你是直接用了同样的方法，比如用Claude提示优化器或者Anthropic提示优化器来制作的吗？
Did you just use did you just do the same thing where you you use, like, a the Claude prompt improver, the anthropic prompt improver to make that?
或者说是的。
Or yeah.
你是怎么考虑把这些内容组合在一起的？
Why how did you think about putting that together?
是的。
Yeah.
这是复利效应的一部分。
This is part of like the compounding effect.
这就像有一个能产生很多结果的想法。
It's like having an idea that has like a lot of a lot of outcomes.
这是Nitesh发给我的。
So this was what Nitesh sent me.
他说，我们刚刚实现了AGI。
He said, we just got AGI.
它已经交付了，我们可以编写软件了。
It got delivered and we can write software.
这是你的初始提示，还挺有趣的。
This was your initial prompt, which is kind of fun.
它非常具有戏剧性。
Like, it's very dramatic.
然后ChatGPT说，我准备好了。
And then ChatGPT said, I'm ready.
好的。
Okay.
所以现在就这么做。
So now do this.
我当时就说，好的。
And I was like, okay.
是的。
Yeah.
没问题。
That's fine.
没问题。
That's fine.
但你知道Anthropic控制台的提示优化器吗？
But like, do you know, the Anthropic Console prompt improver?
你可能会问，那是什么？
You're like, what is that?
嗯，不知道的人听好了，他们改了。
Well, anyone that doesn't know, this is the Oh, they changed it.
这是控制台吗？
Is this the console?
是的，没错。
Yeah, it is.
哦，确实是。
Oh yeah, it is.
好的。
Okay.
他们做了一点修改，但这样很棒，因为你只需粘贴一个提示词之类的内容，然后可以说‘我们在思考’，接着生成，它就会自动优化提示词。
They changed it a little bit, but this is great because basically you paste in a prompt or something like that and you can say, yeah, we're thinking and you can generate and it will improve the prompt automatically.
你会想，它能有多好？
And you think like, how good can it be?
它相当不错，因为使用起来也非常低门槛。
It's pretty good because it's also very low friction.
所以，只需花一分钟看看有没有结果、是否有效，就非常容易。
So like it's very easy to just take a minute to see if something comes out, if it works.
如果不行，就删掉。
If it doesn't work, delete it.
没关系。
Doesn't matter.
我们当时只是随便聊聊，想着要列出30个研究任务，所以最好先写个提示语。
We were just jamming, and we were like, well, we're going to come up with 30 research tasks, so, like, we better have a prompt.
于是我直接复制了这个提示语。
So I just copied this prompt.
然后这就成了文档。
And that became the document.
是的。
Yeah.
好。
Okay.
粘贴到这里，修改参数，然后通过输入斜杠来触发这些指令，我们这里有两个自定义提示。
Into here and change the arguments, and then you can trigger those in clause by doing slash, and we have these two custom prompts here.
展开剩余字幕（还有 416 条）
然后我认为这让我对你们所说的‘工程复利’有了更清晰的理解，因为这告诉我，你们首先花时间构建了一个能自动生成其他提示的提示，因为这些研究文档本质上就是用于 Cloud Code 的提示。
And then I think that actually gives me a much better idea of, like, what you mean by compounding engineering, because what it says to me is what you did first is spent time building a prompt that effectively builds other prompts, because those research documents are effectively prompts for Cloud Code.
所以现在你们有了一个能生成提示的提示，每次想要开发新功能时，需要指定的内容就更少了。
And so now that you have a prompt that builds prompts, every time you want to make a new feature, you have to specify less.
你只需要说一下这个小功能，它就会自动去研究并扩展成一份完整的文档，而以前每次做功能时，都得先说：‘我想让你先做研究’，然后‘再思考所有可能的边界情况’，或者‘我喜欢什么样的实现方式’等等。
Like, you just say the little feature, and then it'll go do the research to build it out into a big document, versus before, every single time you have to do a feature, have to say, at first, I want you to research it, and then I want you to think through all these different corner cases, or the ways that I like things built or whatever.
我觉得这太酷了。
I think that's so cool.
而且特别有趣的是，它在我们说话的时候其实一直在运行，这完全是一种不同的编程方式。
And what's also really interesting to point out is it's working while we've been talking, and that's just a different way to code.
我们上周或者前周还在电话里一起测试这个功能，我当时就在通话过程中上线了一个功能，而我根本没碰过代码库，这简直有点疯狂，但确实发生了。
We were, you know, we were on the phone together, like, last week or the week before, and we were testing this out together, and I shipped a feature that went to prod while we were talking, which I'm not in the code base at all, so it's like kind of crazy that that actually happened.
这就像一种更社交化的编程方式——我们现在一边聊天一边构建东西，这在以前是无法想象的。
And it's it's like it's a kind of more social way to code, like we're coding right now building stuff, which was not possible before.
你想要一份能真正帮你拿下大单的演示文稿吗？
Do you want a winning deck that'll actually help you land big deals?
Pitch 将设计、协作和交付全部整合在一个平台上。
Pitch combines design, collaboration, and delivery all in one platform.
使用 Pitch，不再有版本混乱，也不再有非官方风格的幻灯片。
With Pitch, there's no version chaos and no off brand slides.
它包含了创建和展示专业、精美演示文稿所需的一切功能。
It's everything that you need to create and present professional, beautiful decks.
Pitch 的新 AI 功能可让您立即为任何受众重写幻灯片文字。
Pitch's new AI actions let you instantly rewrite slide text for any audience.
这些功能可帮助您保持品牌语气，自动生成幻灯片摘要和演讲者备注，甚至无需从头开始即可增强视觉效果。
They let you maintain brand tone and automatically generate slide summaries and speaker notes, and they even let you enhance visuals without starting over.
Pitch 管理从设计到交付的整个演示流程，并且在不牺牲设计质量的前提下完成这一切。
Pitch manages your entire presentation workflow from design to delivery, And it does all of this without compromising on design quality.
它具备基于云的协作、品牌素材库、交互式嵌入和智能缩放工具。
It features cloud based collaboration, brand libraries, interactive embeds, and smart scaling tools.
借助 Pitch，您创建的每一张幻灯片都能保持品牌一致性，得益于自定义模板和品牌语气。
With Pitch, every slide you create stays on brand, thanks to custom templates and brand tones.
你分享的每一份演示文稿都会追踪互动情况，让你知道潜在客户是否已打开。
And every presentation you share tracks engagement, so you know whether prospects have opened it.
如果你面对新受众，AI会立即重写你的信息。
If you have a new audience, AI rewrites your messaging instantly.
你想追踪哪些幻灯片引起了共鸣吗？
Do you wanna track which slides resonated?
互动分析会告诉你答案。
Engagement analytics show you.
如果客户需要为他们的交易创建专属空间，可以建立自定义演示室。
If your client needs a dedicated space for their deal, create a custom pitch room.
演示平台会适应你实际的工作方式。
The pitch platform adapts to how you actually work.
当你准备演示时，可以通过实时链接或自定义房间进行分享。
When you're ready to present, share via live links or custom rooms.
Pitch负责 hosting、追踪和专业润色。
Pitch handles the hosting, the tracking, and the professional polish.
你准备好在一个平台上设计、协作并促成交易了吗？
Are you ready to design, collaborate, and close deals all on one platform?
立即在 pitch.com 免费开始创建，几分钟内即可为你的团队完成设置。
Start creating for free at pitch.com and get your team set up in minutes.
那就是 pitch.com。
That's pitch.com.
今天就开始你的免费试用吧。
Start your free trial today.
现在回到本期节目。
And now back to the episode.
是的。
Yeah.
当然。
Absolutely.
哦，对了，就在我们聊天的时候，我们做了研究并创建了这个内容，挺棒的。
And oh, so it's while we were talking, we did the research and we created this issue, which is cool.
我们当时大概有六七个想法同时进行，因为我们一有新点子就马上行动。
And we had, I think, six or seven running at the same time because we were just like, new idea, let's go.
新点子，马上干。
New idea, let's go.
我们还做了另一件事：收集用户反馈，阅读邮件，能想到的所有渠道我们都翻了一遍，然后一起头脑风暴。
And what we also did, we went through user feedback, we read emails, we just everything we could, we gathered and we were just like brainstorming.
这真的特别有趣，因为当你处在头脑风暴的状态时，可以随时启动智能代理，看看它们能产生什么想法，过段时间再回来复盘。
And it's really fun because if you're in this brainstorming place, you can just kick off agents and see what comes up, what they come up with and take another time to then review.
我们还很认同你的观点：一起在电话里做这件事真的很有趣，因为真正的灵感往往就在这里诞生。
So what we do also is, like to agree with you on like, it's really fun to do this together on a call because that's where magic happens.
但这里仍然需要人工审核，因为我们发现必须亲自查看，判断是否合理、有没有遗漏——这需要品味、经验和直觉。
And there is still a human review step here because we found that we want to look at it, see if it makes sense, if anything is missing like this is having taste, experience, intuition like this.
这就是我之前解决的那个邮件发送失败的bug。
This the bug I solved earlier with the email not going out.
娜塔莎也对她的代码做了同样的事，但没给出正确的结果。
Natasha did the same with his clothes code, but it didn't give the right answer.
是的。
Yeah.
所以，仍然需要人类的直觉和触感。
So so there is like there is still like a human touch of intuition.
就像我提到的，查看历史记录确实引导它走向了正确的方向。
Like I hinted at look at the history and that actually made it think into the right direction.
然后Tesh没有查看历史记录，结果它说：‘一切正常。’
And then Tesh didn't add look at the history and then it said, no, everything works fine.
所以，直觉仍然存在，这仍然是一种技能。
So there is still like intuition and like It's still a skill.
这仍然是一种技能。
It's still a skill.
这无疑是一种技能。
It is a skill for sure.
是的。
Yeah.
这可不是嗯。
It's not yeah.
这绝对是一种技能。
It's absolutely a skill.
没有哪种魔法提示能包办一切。
There's no magic prompts that does everything.
关键是用对方法，充分发挥它的优势。
Like, it is about using it the right way and using it to its strengths for sure.
是的。
Yeah.
对。
Yeah.
尼特什，你对这一切感觉如何？
Nitesh, how have you found this all?
因为我知道，基兰是位资深的 Rails 专家，编程能力超强，而我觉得你才刚开始编程之路。
Because I know, you know, Kieran is, like, a long time, like, Rails, like, expert person who's just, an incredible programmer, and I think you're a little bit earlier in your programming journey.
那么，从零开始在Quora工作，并以这种方式上手，感觉怎么样？
So what has that been like to come to every start working on Quora and start working on it in this way?
是的。
Yeah.
不。
No.
这真是一次极其开阔眼界的经历，说实话，我以前对编程的理解是：两年前ChatGPT刚出来时，我觉得这正是我自学编程、开发那个我一直想做的SaaS应用的绝佳机会。
This has been, like, incredibly eye opening, I would say, because honestly, like my experience with programming is that two years ago when ChatGPT came out, I thought, okay, now it's perfect for me to teach myself programming and build that SaaS application and that I always wanted to.
所以我从第一天起就用ChatGPT自学编程。
So I taught myself programming using ChatGPT from the very first day.
我经历了所有的转变。
So I have gone through all the transitions.
我从ChatGPT开始，后来Cursor推出后，我就转用了Cursor的工作流程。
I went from ChatGPT, and then when Cursor came out, I shifted to workflow to Cursor.
再后来，我们转向了Windsurf的卡片面板。
And then the Windsurf card panel, we shifted into Windsurf.
而且，我一直觉得自己站在前沿。
And, you know, I was always thinking like, okay, I am at the forefront.
我不认识任何朋友像我这样大量使用人工智能走在前沿。
I don't know any of my friends who are doing so much with AI at the forefront.
然后我加入了Avery，开始和Kieran一起工作。
And then I joined Avery and start working with Kieran.
Kieran的水平完全不一样。
And Kieran is at a whole other level.
在我们的会议中，他直接在写代码。
In our meetings, he's like, now he's writing code.
他从不打字。
He's never typing.
他总是对着电脑说话。
He's always speaking into the, computer.
所以我想，好吧，我得把这种方式纳入我的工作流程。
And, so I was like, Okay, I need to log that into the workflow.
后来当Claude Code发布时，Kieran实际上推动我使用它。
And then even when Claude Code came out, Kieran actually pushed me into using it.
很明显，现在这已经成为编程的方式。
And clearly, it is now the way to program.
我和Kiran两个人，最近三周左右根本没碰过WinServe或Cursor。
Like, me and Kiran, both of us, like, we haven't even touched WinServe or Cursor in the last, like, three weeks or so.
即使我们偶尔用一下，通常也只是为了读点东西。
Or even if we do touch it, like, it's it's usually just because we wanna read something.
我们用它纯粹是因为电脑上没有Visual Studio Code。
It's it's basically like we're using it because we don't have Versus Code on our computer.
不管它是旧版的Visual Studio Code还是Cluster，其实都没关系。
Like, it wouldn't matter if it was Versus Code, like the older Versus Code or Cluster, and stuff.
因为现在所有的AI功能都集中在Cloud Code上，能处在这种环境中真的很有趣——整个编码领域每三个月就彻底改变一次。
Because all the AI stuff is happening with Cloud Code now, and it's really fun to, you know, have be in this position where the entire coding landscape just changes completely every three months.
你会意识到，其实没人真正站在前沿。
And you realize, like, nobody's at the forefront.
我得说，我很羡慕你在Chattypu T刚推出时就开始学编程，因为我二十年前是靠书本学编程的。
I gotta say, I'm jealous of you learning to code right when Chattypu T came out because I learned to code from books, like, twenty years ago.
B h b四入门。
B h b four for dummies.
是的。
Yeah.
比如《24天学会Basic》这样的书。
Like, basic for learn basic in twenty four days.
像Sam的《自学者Basic》之类的。
Like, Sam's teacher self basic or whatever.
对。
Yeah.
Delphi 5。
Delphi five.
对。
Yeah.
而且，而且你这么说真的很好笑，我本来以为自己算是AI编程的前沿人物了，但后来我加入了Every，开始和基兰一起工作，这让我想起《星球大战》前传第一部里有个场景，他们 underwater，被一只海怪攻击，眼看就要死了。
And and and also, it's so funny for you to to say, like, I thought I was I thought I was sort of at the at the forefront of AI coding, and then I joined Every and started working with Kieran because it just reminds me of I don't know if like, there's a scene in Star Wars, the prequel episode one where, like, they're they're under the water and and they're, like, being attacked by a sea monster, and it looks like they're gonna die.
然后另一只更大的海怪突然出现，直接把那只攻击他们的海怪吞了，魁刚说：总会有更大的鱼。
And then another bigger sea monster comes out and just, like, eats the eats the the one that's killing them, and Qui Gon is like, there's always a bigger fish.
总会有更大的鱼。
There's always a bigger fish.
而没错，基兰就是那只更大的鱼。
And, yeah, Kieran Kieran is the bigger fish.
但我也有同样的感觉。
But I feel I feel the same.
就像你对我说的那样，但我自己根本不知道自己在做什么。
Like like you say that about me, but I'm like, I'm I have no idea what I'm doing.
我得赶紧跟上，现在落得太远了。
Like, I need to, like, I'm running behind.
我们还得做上百万件事才行。
We need to do like a million more things.
这就是现状的现实。
So that's just the reality of the landscape.
就像总是有更多东西，但关键在于练习。
Like there is always more, but it's really about practice.
你应该练习使用AI。
Like you should practice using AI.
你应该每天挑战自己。
You should push yourself every day.
如果你不这么做，就会错过很多酷炫的东西。
If you don't, like, you'll miss very cool stuff.
是的。
Yeah.
那么，我认为大家可能都很好奇，无论是我个人还是观众，这个问题有什么弊端呢？
Well, what are I I I guess I'm curious, like, personally, and also for people in the audience, like, what are the problems with this?
对吧？
Right?
所以，基本上听起来你正在转向一种不直接编写代码的编程方式。
So, basically, it sounds like you're moving to a form of coding where you don't touch the code.
你处于更高的一个层级。
You're one level above.
那么，这种方式会带来哪些问题？你又是如何解决这些问题的？
And so what are the problems that come up with that, and how are you solving them?
为了确保一切顺利，你需要引入哪些新的工程实践？
What are the new engineering practices that you need to incorporate in order to make sure that things go well?
对我来说，最重要的一个领悟就是，我总是反复想到这一点，尤其是在使用短代码时。
For me, like, the the most important realization for me has been, like, this thing that I always keep going back to, especially with shortcode.
我读过一本管理类书籍《高产出管理》，那是英特尔前CEO五十年前写的。
I read this in that management book, that high output management, which the Intel CEO wrote, like, fifty years ago.
在第一章中，他提到，在任何生产流程中，都应在最低价值阶段解决任何问题。
And in the first chapter, he mentioned something like how in any production process, you should fix any problem at the lowest value stage.
我无法停止思考这句话，因为AI和低代码现在能为我们完成这么多事情，因此专注于最前端的环节变得至关重要。
And I just can't stop thinking about that statement because AI and plot code can now do so many things for us, it has become really important to focus on the earliest part of things.
我的意思是，当我们看到这种情况时，比如使用基兰刚刚展示的工作流来创建一个非常详细的 GitHub 问题时，很容易就想再开一个 Cloud Code，让它直接去处理这个 GitHub 问题并修复它。
So what I mean by that is when when we see that, you know, when we are using the workflow that Kieran just showed to create a GitHub, like a very detailed GitHub issue, then it's very tempting to like start another Cloud Code to ask it to just, hey, go now, work on this GitHub issue and fix it.
但这样做其实会有问题，因为克莱德在那个问题中给出的计划，可能并不是你真正想走的方向。
But that's actually gonna be a problem because there are chances that, you know, the the plan that Claude was able to give in that issue, it wasn't the direction that you wanted to go.
你希望在让克莱德去实施解决方案之前就发现这个问题，然后及时修正。
And you want to catch that before you ask Claude to go and implement the solution, and then you wanna fix it over there.
这完全说得通。
That makes perfect sense.
我非常非常欣赏这个想法。
I really, really like that idea.
让我想到的是，所有这些东西就像一根杠杆，你离支点越远，力量越大，但同时也更容易朝错误的方向发力。
The thing it reminds me of is just it's like all this stuff is like it's like a lever, and the further out you get on the lever, the more power you have, but also the more power you have to go in the wrong direction.
每一点微小的偏差，到了末端都会造成巨大差异，所以尽早发现问题，我认为这才是确保你不会失控的关键——虽然我这个杠杆的比喻可能有点破绽，但你懂我的意思吧？
Like, every little inch makes a big difference at the end, And and so trying to catch it earlier, I think, is the thing that makes sure that you're not shooting off into space or Liz Lever metaphor is totally breaking, but, like, you know what I mean?
比如，如果你把火箭对准月球，哪怕只偏了一英寸，最终也会差上数千英里。
Like, if you if you point a rocket at the moon, like, one inch means thousands of miles of difference.
所以我认为AI领域也是如此，这其实对我来说是个很好的教训，因为我总是倾向于急于跳过规划阶段。
And so I guess the same thing is true with AI stuff, and I think that's actually a good lesson for me because I I tend to wanna, like, rush through the planning stuff.
就像，对我来说，很难去盯着那样的文档——就是我们写下的那些内容——并专注其中。
It's like, it's just hard for me to, like, look at a document like that, like the thing that's that caught us writing and concentrate on it.
那你们是怎么看待这个问题的呢？
So how how have you guys found that?
是的。
Yeah.
大多数时候读这些内容都挺无聊的。
It's it's kind of boring to read most of the time.
但你可以让它变得更有趣。
But you can make it more fun.
比如你可以要求极简、极简的内容。
Like you can say just like minimal, minimal.
这太多了。
This is too much.
但问题是，这样又会漏掉一些东西。
Just but but then the thing is then it misses things again.
所以这其实很重要。
So it's actually important.
所以对于代码，我喜欢它聚焦在用户故事上，或者提出问题并回答它们。
So for code, I like it to focus on user stories or like asking questions and answering them.
比如说，一个优秀的产品经理会问哪些关于这个问题的问题，我们应该考虑一下，并给出两个选项？
So let's say like, hey, what are some questions a good PM would ask about this, like that we should consider and give like two options?
这样读起来比‘第一周做这个，第二周做那个’有趣多了。
Like that's it's more fun to read that than like week one, we'll do this week two, we'll do that.
PRD本来就很无聊，但你可以让它稍微有趣一点，多举些例子，或者调整一下研究的呈现方式。
Like that's it's like PRDs are boring and you can make them a little bit more fun or give more examples or like you can shape that research.
这通常就是我们在人工审查阶段所做的。
And that's normally what we do in the in the human review step.
比如，我们有没有发现什么警示信号？
It's like, do we see any red flags?
我们需要添加更多内容吗？
Do we need more stuff to be added?
因为这样能节省很多时间。
Because it will save so much time.
是的。
Yeah.
这让我想起我们在公司其他部门发现的一个情况。
That That actually reminds me of something that we're finding in another part of the business.
丹尼，他之前上过这个节目，是Spiral的总经理，而在Spiral内部，我们正在开发一个写作助手。
So Danny, who's been on this show, is the GM of Spiral, and inside of Spiral, we're building a writing agent.
你可以把它想象成类似热代码的东西，但专门用于写作任务。
So you can think of it sort of like hot code, but specifically for writing tasks.
我觉得这其中有一些相似之处：有时候你希望这个写作助手能切换到访谈模式，去深入了解你是谁、你需要什么，而不是直接抛出一堆内容让你去阅读。
And I think there's something similar about that, where sometimes you want that writing agent to shift into, like, an interview mode where it, like, tries to understand more about what who you are and what you want, rather than just, like, spitting out a bunch of stuff that you then you have to read through.
听起来，ClaudeCode或这类编码工作流程中可能缺少了某种东西——与其让你去读一份冗长的文档，不如让它主动向你提问，这样它输出的结果更可能符合你的需求，而你就不必重做整个过程了。
And it sounds like there's maybe something missing here in ClaudeCode or these sort of coding work coding workflows where it would be really nice, instead of having to read that, like, long document, it's finding ways to, like, ask you questions so that the thing it outputs is more likely to be right without you having to redo the whole thing.
是的
Yeah.
当然
Absolutely.
这是一个关于此处自定义命令点击的有趣想法。
That's an interesting idea for a custom command click here.
我们绝对应该试试。
We should totally try that.
是的
Yeah.
肯定的
For sure.
这确实是我们应该自动化并改进的事情，毫无疑问。
Like, this is something we should automate and make better, like, for sure.
同时，它因为能访问你的代码库和风格，所以了解很多。
And at the same time, it it knows a lot because it has access to your code base and your style.
而且，这真的非常强大。
And like, that's that's very powerful.
所以，你有代码库，它实际上做得相当不错。
So like like you have the code base and it's actually pretty good doing it.
我觉得，除了刚开始就把它做得很好之外，传统的测试和检查也非常重要。
Like, I think in addition to like making it very good at the beginning, I think just boring traditional tests and evils are very important as well.
因为你怎么知道你做的东西真的在工作呢？
Because how do you know what you did is actually working?
你可以打开控制台手动点击测试，但为什么要这么做呢？
Well, you can open a console and click through it, but like, why?
直接让它自动测试吧。
Just have it tested.
为它写一个测试。
Write a test for it.
哪怕只是最起码的测试。
Like just just the bare minimum.
烟雾测试很棒，你只需要看看它是否基本能工作，否则它会做太多事情。
Smoke tests are great where you just see does does it kind of work because otherwise it does way too much.
但这是让它自行迭代和修复问题的好方法。
But it's a very good way to have it iterate and fix things by itself.
我们还没怎么尝试过，但我们用过一个Figma MCP，说‘从Figma实现这个’。
And we haven't tried it as much yet, but we used a Figma MCP where we say, hey, implement this from Figma.
现在你可以让Puppeteer为移动版本截一张图，然后比较两张图。
And then now there is like you can have Puppeteer take a screenshot for a mobile version and then say, compare the two.
我们其实还没怎么试过，但我们想多尝试这类方法。
Like, we haven't really tried it out, but like we want to try more of that out.
所以这些检查和测试通常是手动进行的。
So there are these checks in place, tests in place that you normally do manually.
对提示词也是如此，就像为提示词做评估。
And the same for prompts like evils for prompts.
所以我把评估看作是为代码写测试。
So I kind of think of an eval as like writing a test for a code.
评估就是对提示词的测试。
An eval is a test for a prompt.
上周我还看到，Cloth Codes运行了一次评估，结果发现它在10次中有4次失败。
And what I've seen last week as well, I had cloth codes run an eval and then say, actually it fails four out of 10 times.
我说：运行10次。
I said, run it 10 times.
它每次都通过吗？
Does it always pass?
不，有4次没通过。
No, four times doesn't.
我说：看看输出结果。
I said, oh, look at the output.
为什么它没有调用那个工具呢？
Well, like, why didn't it call that tool?
那是一个很酷的工具，叫test。
It was a cool tool tool called test.
它说，哦，是不够具体。
And it says, oh, yeah, it wasn't specific enough.
我说，好吧，继续调整提示，直到它能始终稳定通过。
And I said, okay, just keep going and change the prompt until it's passing consistently all the time.
它真的做到了。
And it did it.
就像我下楼喝了杯咖啡，再上楼时，事情就解决了。
It's like I just walked downstairs, got a coffee, walked up and that was it.
所以评估也非常强大，因为它们能告诉你提示是否有效。
So evals are also very powerful because they will tell you if a prompt works.
就像写代码一样，测试能证明你的代码是有效的。
And similar to writing code, tests says your code works.
因此，更多地采用这些看似枯燥的传统方法也非常有效。
So leaning into those more boring traditional ways is also very powerful.
这说得通。
That makes sense.
我有个想法。
I have a thought.
我觉得其中一件特别的事情是，Nitesh，你也面临这种情况，告诉我我是不是错了，但我觉得你，Karen，特别之处在于你总是测试一切。
And because one of the things I think is really special, and I think, Nitesh, you're in this boat too, so tell me if I'm wrong, but one of the things I think is really special about you, Karen, is that you just test everything.
所以，你测试过每一个代理。
So, like, you've tested every single agent.
Nitesh，你也用过很多代理吗？
Nitesh, have you have you used, like, a lot of the agents as well?
没有。
No.
那是Kieran。
That's Kieran.
好的。
Okay.
我觉得我们还是可以做到的。
Well, I I think we could I think we could still do this.
我觉得这会挺有趣的。
I think it'd be kinda fun.
我想花五分钟和基兰一起，对这些代理进行从S级到F级的排名。
I wanna spend five minutes with Kieran doing a s tier through f tier ranking of agents.
所以我要做的就是共享屏幕，我会报出一个代理的名字，然后你告诉我它排在第几级。
And so what I'm gonna do is I'm gonna share my screen, and I'm gonna I'll call out an agent, and then you tell me where it ranks.
你愿意参与吗？
Are are you game?
好的。
Yeah.
我们开始吧。
Let's do it.
好的。
Okay.
太好了。
Cool.
我们来看看。
Let's see.
我想想，咱们先来排个 Cursor 吧。
Does does I guess cur let's do let's do cursor.
是的。
Yeah.
这挺有趣的，因为 Cursor 到底是什么？
So it it's fun because cursor like, what cursor?
是 Cloth Four 吗？
Is it cloth four?
是 Max 吗？
Is it max?
是
Is it
Cursor 在最佳设置下。
Cursor on the best possible settings.
它是后台代理吗，还是说那个好吧。
And is it the background agent or is it the the okay.
Cursor 传统最佳设置关闭。
Cursor traditional best possible setting close.
这正是 Cursor 和 Windsurf 让人困惑的地方。
Like, that's the confusing part about Cursor and Windsurf.
就像，它有上百万个版本。
Like, there are, like, a million versions of it.
那为什么你不直接用最好的版本呢？
And like, why don't you just have the best version?
这正是我喜欢某些代理的原因。
And that's what I love about certain agents.
它们直接说：看，这就是最好的代理。
They just say, look, this is the best agent.
所以这就是为什么它不会是最好的。
So that's why it wouldn't be the best.
我会说
I would say
啊。
a.
啊？
A?
好的。
Okay.
Cursor是
Cursor is
Cursor与Claude四代配合得很好。
Cursor is very good with Claude four.
明白了。
Alright.
Windsurf。
Windsurf.
C，因为他们没有Clue 4。
C, because they don't have clot four.
这太荒谬了，因为三周前他们还是A，现在却不是了。
It's ridiculous because three three weeks ago, they would be A, and now now they're not.
哇。
Wow.
哇。
Wow.
好的。
Okay.
因为我几个月前从Windsurf切换到了Cursor，或者从Cursor切换到了Windsurf，但后来我又换回来了。
Because I I switched from Windsurf to or from Cursor to Windsurf, like, a few months back, but I switched back.
好的。
Okay.
所以Windsurf是C，Cursor是A。
So we've got Windsurf is a Windsurf is a c, Cursor is an a.
我们来看看。
Let's see.
Devin。
Devin.
这是一个b。
It's a b.
b。
B.
为什么？
Why?
它不像其他工具那样集成得那么好。
It's like, it's not as integrated.
设置起来有点麻烦，代码质量也不如Cursor或Cloak Code那样完善。
It's a little bit hard to set up and the code quality is like, it's not as well rounded as cursor or Cloak Code.
我不确定他们是否用Cloth做后台，但它的易用性确实不如其他工具。
I don't know if they use Cloth for the background, but like, it's not as usable as the others.
查理。
Charlie.
查理主要用于代码审查。
Charlie is like for code reviews.
所以我们主要用查理做代码审查。
So we use Charlie for code reviews mostly.
所以我并没有怎么把它当助手用。
So I haven't really used it as an agent as much.
我觉得查理作为助手是B级，但作为代码审查工具是A级。
I think Charlie as an agent is B, but it's A as a code reviewer.
我真的很喜欢它做的代码审查。
Like I really like the code reviews it does.
这挺有意思的。
So that's interesting.
它在某方面真的很出色。
Like it it's really good at something.
那周五呢？
And then what about Friday?
我把周五排在Cursor前面，可能在S和A之间。
I put Friday higher than cursor, maybe between S and A.
有趣的是，他们甚至还没用Clot four呢。
And it's funny because they don't even use Clot four yet.
他们还在努力想办法让它真正好用。
They're still they're still working on how they like really make it work well.
它是3.7，但我喜欢它的地方在于，它和Clothcote完全不同，但Friday的工作方式非常有主见。
It's 3.7, but like why I like it there, it's definitely different than Clothcote, but Friday has a very opinionated way of working.
我很欣赏他们的观点，它确实运行得很好，而且能自动完成任务。
And I love their opinions and it really works well and it just does it.
比如你提出一个问题，它们会制定计划，你批准后，它就会自动执行。
Like you give an issue, they make a plan, you approve and it does it.
它会创建一个拉取请求。
It creates a pull request.
我见过它完成了一些我用冷代码做不到的事情。
And I've seen it do this stuff that I couldn't do with cold codes.
比如，实现这个Figma设计。
Like, for example, implement this Figma design.
它只是展示了一个给助手的Figma设计。
It just one showed a Figma design for the assistant.
我见过好几次这样的时刻，它做了些让我惊叹的事情，仿佛我尝到了未来的味道，这真的很独特。
And and I've seen moments where more multiple moments like that, where it did things are like, wow, okay, this I taste the future, which is really unique.
而且它也是一个小团队。
And it's a small team as well.
真的很棒。
So really cool.
Codex。
Codex.
这对我来说是B。
This is b for me.
好的。
Alright.
Codex 是 b。
Codex is a b.
Copilot。
Copilot.
我还没用过 Copilot。
I haven't used Copilot.
你从来没用过 GitHub Copilot？
You never used GitHub Copilot?
没有。
No.
我的意思是，三年前用过，但我不用了。
I mean, used it three years ago, but I like no.
我，好吧。
I I okay.
咱们公平一点。
Let's be fair.
我大概半年前试过，但一秒钟后我就不再用了。
I tried it maybe a half a year ago and after one second I stopped using it.
那你对它的评价如何？
So where do where do you rank it?
D级。
D.
它不具备自主性，但我也确实应该试试新版。
Like, it was not agentic, but but, I mean, I should try the new version for sure.
我们还没试过。
We have not tried yeah.
我们没试过自主型Copilot，所以这么说也不完全公平，但好吧。
We've we haven't tried the the agentic copilot, so that's that's not totally fair, but okay.
我们还漏了什么吗？
Are we missing anything?
我觉得我们确实漏掉了。
I feel like we are.
Claude。
Claude.
Claude的代码。
Claude's code.
当然，Claude的代码。
Well, obviously, Claude code.
但我假设它是S级的，宝贝。
But I I assume s tier, baby.
我们
We
我们也有Factory。
have factory as well.
哦，是的。
Oh, yeah.
你觉得工厂这个工具该怎么评价？
What's how where do you rank factory?
这挺有意思的。
It's interesting.
工厂在某些方面确实比其他工具更好，但并不符合我的风格。
Factory with certain things is, like, better than any others, but it's not my style.
工厂更适合那些非常极客、追求极致代码表现的企业级用户。
It's like factory is for more enterprise y people that are very nerdy and want like absolute bangers of code.
它确实不错，比如支持多仓库之类的功能。
And it's actually good, like multi repo, stuff like that.
因为它既能在网页上使用，也能本地运行，所以用起来有点复杂。
It's a little bit hard to use because it's on the web, but also local.
所以我给它打B分。
So I rate it b.
可能比Codex和DevIn稍微低一点。
Maybe a little bit below Codex and and dev in.
是的。
Yeah.
但它的用途确实是存在的。
But it's like, there is a use for it for sure.
有什么事情正在发生。
There's something going on.
那里确实有不错的东西，但可能不适合我们。
There's something good there, but it's maybe not not for us.
这不是我的风格。
It's not my thing.
是的。
Yeah.
是的。
Yeah.
AMP也是。
AMP also.
AMP。
AMP.
那是什么？
What's that?
AMP。
AMP.
A m p。
A m p.
AMP。
AMP.
哦，AMP。
Oh, AMP.
是的。
Yeah.
是的。
Yeah.
抱歉。
Sorry.
那怎么样？
How's that?
是的。
Yeah.
会把它放在外套里面的S级位置，介于哇哦之间。
Would put it S tier under cloth coat between Woah.
另一个S级。
Another S tier.
对。
Yes.
好的。
Alright.
为什么？
Why?
它在完成工作方面非常出色。
It's it's very good at just getting work done.
经济效益也不错。
The economics are pretty good.
已经有很好的工具了。
Good tools already.
就像人们用那个工具来构建它一样。
Like people people use that tool that build it.
他们在内部使用。
They're dogfooding.
你可以从Clothcodes和Ant的开发者身上感受到，他们热爱智能体，正在打造最好的东西，并不断尝试新事物。
Like you can feel from Clothcodes and Ant are developers that love agents and they're just building the best thing and they're trying new things out.
所以，是的，就是这样。
So, yeah, that's it.
我们来看看。
Let's see.
这就像基兰是大鱼一样。
This is exactly like Kieran as the big fish.
我的意思是，但它确实是
I mean, but it's
你确实是。
You are yeah.
你正在把它们串联起来。
You're you're stringing them together.
比如，你同时使用了Cloud Code、Friday和其他工具，这确实是特别酷的地方。
Like, you're using Cloud Code and Friday and and other stuff all at the same time, which is, yeah, the thing that is really cool.
是的。
Yeah.
就像，确实有。
Like, like, there are yeah.
我是这么想的，就像你在为某个职位面试，然后找一个开发者来解决特定的问题。
Like, how I think about it, like, I'm I'm thinking about it more is like you're interviewing for a role and you find a developer to, like, solve a certain problem.
我觉得编码代理也是类似的。
I think it's similar with coding agents.
比如，Friday 现在很擅长做界面工作。
Like, Friday is good at like doing UI now.
所以如果我需要做界面工作，就会找 Friday。
So if I need UI work, will go to Friday.
如果我需要做研究，就去找 ClothesCodes。
If I need to do research, go to ClothesCodes.
而且如果我想进行代码审查，我会用 Charlie。
And yeah, there is a if I want a code review, I use Charlie.
这很有趣，而且这些代理可以协同工作。
Like it's it's fun, and agents work together.
你不需要只依赖一个代理。
You don't need to have one agent.
我们称之为 codes。
We have called codes.
这是因为Charlie在GitHub上工作，所以你只需要@Charlie，它就会对PR进行代码审查。
And that's because Charlie, like, works in GitHub, so you can just, like, CC Charlie, and and Charlie will do the the code review on the PR.
是的。
Yeah.
所以我们使用GitHub、拉取请求和常规的开发流程。
So we use GitHub and pull requests and normal developer flows.
人类可以介入。
Humans can hook in.
因此，我们可以雇佣擅长特定任务的人来审查代码，而闭合代码则负责完成工作。
So we can hire someone that's very good at specific thing and review codes and then close code will just do the work.
但这非常强大，因为我们已经像这样打磨了二十年左右的生态系统，而且它很有效。
But it's very powerful because it is just an ecosystem that we refined over like twenty years or whatever, like, and it works.
所以让我们充分利用这一点。
So let's lean into that.
这可能也是为什么Copilot已经融入其中后表现良好。
And that's probably why Copilot will probably be fine since it's in there already.
是的。
Yeah.
等等，你最近真的这么做了。
Wait, you actually did that recently.
比如，我们在Quora处理过海量邮件，因此遇到一些基础设施问题，你似乎请来了一位真正的专家，以一种特定的代理方式与他们合作，既达到了目标，又减轻了他们的负担。
Like, we had some infrastructure things where, you know, we've handled tons and tons and tons of emails at Quora, so we had some infrastructure issues to work out, and you seemed I think you brought in someone who's like a real expert, and then worked with them in a specific agentic way that you got what you needed from them, but it was less work for them.
是的。
Yeah.
是的。
Yeah.
当时还没有出现任何问题，但我们希望更清楚地掌握最重要事项的交付情况。
So, like, like there was no issue yet, but we wanted more visibility in delivery of like the most important things.
我不太擅长这个，或者说我懂一些，但还是请个专家来吧。
Like, I'm not very good at it or like, I know stuff, but like, let's bring in someone.
我们做的是，进行了一次两小时的对话，并把整个过程录了下来。
And what we did, we just had a conversation like a two hour call and I recorded everything.
最后，我把这些内容输入到Cloth中，说：‘好吧，你能根据这个生成两个资源问题吗？’
And at the end, I just fed that into cloth and say, okay, can you make two issues, resource issues from this?
大约十分钟后，我说：‘好了，这是问题列表。’
And like ten minutes later I said, okay, here are the issues.
你能帮我看一下吗？
Can you review them?
他惊讶地说：‘天啊，这是怎么回事？’
And he was like, holy, what?
这位先生并不是AI怀疑者，但他确实非常擅长自己的工作。
This guy, like he's not an AI skeptic, but he's like, he's he's very good at what he does.
通常他做的那些事，AI还做不到，因为有些事情AI目前还不够擅长。
And normally what he does, AI is not good at yet because like there are things AI is not as good at yet.
但他对这个表现非常印象深刻，并给出了很多很好的反馈意见，供我们进一步改进。
But he was very impressed with it and he had like very good like comments on it to iterate over it.
我们实际上只是通过拥有可以讨论的具体内容，更快地迭代了各种想法。
And like what we basically did, we just iterated more quickly through ideas because we had something to talk about.
然后第二天，当他说到‘人类审阅’的时候，我就说：好了，我们开始吧。
And then I said the next day when he was like, did the human review, like, let's go.
我只是用云代码来实现它。
I just use Cloud Code to implement it.
我们坐下来进行了代码审查。
And we sat down and did the code review.
这加速了原本可能需要两周才能完成的工作，现在几个小时就搞定了，真是太棒了。
So it's like it's just accelerated what would have taken two weeks maybe, is now in like a few hours, which is really cool.
我太喜欢了。
I love it.
好了，事情就是这样。
Well, there you have it.
你已经有了代理人的等级排名。
You've got your tier list of agents.
Claude code 赢得冠军。
Claude code takes the cake.
接下来是AMP，还有GitHub Copilot，不幸的是排在最后，但一旦我们尝试他们的智能代理功能，还有很大的提升空间。
We've got AMP coming up coming up in second, and GitHub Copilot, unfortunately, bringing up the rear, but with room for improvement once we try out their their agentic capabilities.
在今天结束之前，你们还有什么想说或讨论的吗？
Anything else you guys wanna wanna say or talk about before we we end today?
每个人都应该使用Cold Code或试一试。
Everyone should use cold code or try it out.
即使你不是技术人员，也订阅他们的Max或Pro版本。
Even if you're not technical, subscribe for their max or pro plan.
每月只要100美元。
It's only a $100 per month.
你可以无限使用。
You have unlimited access.
如果你对技术有疑虑，觉得很难，其实非常简单。
If you're skeptical about being technical, that it's very easy.
我见过一些人，我有个朋友用了Cursor，我就对他说，直接用Cold Code，更好用。
And I've seen people, a friend of mine, he used cursor and I said, just use clothes code, it's better.
他问：还能好到哪里去？
He was like, how much better can it be?
他说：是的，确实更好。
And he said, yes, it's better.
他把他用 Cursor 做的所有东西都重新用 Cold Code 重做了一遍。
And he rebuilt everything he did with cursor vibe coded into close code.
他说：是的，这太棒了。
And he's like, yeah, this is great.
他感受到了下一步，你们每个人都应该试试，并真正推动自己的工具。
He felt that next step and you should everyone should try it and really push push their tools.
对。
Yeah.
尼特什，还有什么建议吗？
Nitesh, any other words of wisdom?
一定要在最基础的阶段检查 AI 的工作。
Just be sure to check the AI's work at the lowest starting stage.
你希望尽早发现这些问题。
You wanna catch those problems early.
是的。
Yeah.
这真是个好建议。
That's that's a great one.
另外，使用 Quora。
And, also, use Quora.
Quora.computer。
Quora.computer.
去看看吧。
Check it out.
它非常棒。
It's pretty awesome.
我们一直在向你推送各种内容。
We're shipping you things all the time.
感谢你们两位的到来。
Thank you both for coming on.
这真是莫大的荣幸。
This is a true pleasure.
我非常期待看到你们在未来几个月里还会创造出什么新东西，我们很快再聊。
I cannot wait to see what else you cook up over the next couple months, and we'll talk soon.
谢谢。
Thank you.
非常感谢。
Thank you so much.
天哪，各位。
Oh my gosh, folks.
你们绝对必须点击点赞按钮并订阅AI and I频道。
You absolutely positively have to smash that like button and subscribe to AI and I.
为什么？
Why?
因为这个节目是卓越的典范。
Because this show is the epitome of awesomeness.
这就像在后院发现了一个宝箱，但里面装的不是黄金，而是关于聊天GPT的纯粹无杂质的知识炸弹。
It's like finding a treasure chest in your backyard, but instead of gold, it's filled with pure unadulterated knowledge bombs about chat GPT.
每一集都是一场情感、洞见和欢笑的过山车，让你坐立不安，渴望更多。
Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat craving for more.
这不仅仅是一档节目。
It's not just a show.
这是一场与丹·希珀共同驾驶宇宙飞船驶向未来的旅程。
It's a journey into the future with Dan Shipper as the captain of the spaceship.
所以，善待自己吧。
So do yourself a favor.
点赞、订阅，并系好安全带，准备迎接你人生中最精彩的旅程。
Hit like, smash subscribe, and strap in for the ride of your life.
现在，不多说了，我只想说，丹，我完全无可救药地爱上了你。
And now without any further ado, let me just say, Dan, I'm absolutely hopelessly in love with you.
关于 Bayt 播客
Bayt 提供中文+原文双语音频和字幕，帮助你打破语言障碍，轻松听懂全球优质播客。