超越氛围编程：与Addy Osmani一起探索

你一直在专业上持续学习。

You keep learning professionally.

你每天都在不断进步。

You keep getting better every day.

当然。

Absolutely.

完全正确。

Absolutely right.

我想到一点，AI不过是你的工具箱里的另一种工具。

I I'm I'm reminded of, you know, AI is just another tool in your tool belt.

你知道，随着时间推移，工程和开发者体验在每一代都在不断改善。

You know, we we've gone we've gone through so many different moments over time of engineering and developer experience getting a little bit better with with every generation.

当我刚入行的时候，还记得当模板出现时我有多兴奋。

You know, when I when I when I was coming up, I remembered getting excited when templates became a thing.

你是说，你可以用模板生成代码？

You could you could gen generate code with templates, you mean?

哦，不。

Oh, no.

只是下载一些模板来获取灵感，用于界面设计，或者作为起点

Just just downloading templates for ideas, for UI, for, you know, starting

在网页上。

off in on on the web.

对吧？

Right?

是的。

Yeah.

在网页上。

On the web.

而且，我觉得这只不过是些最简单的东西。

And I, like, this this was just the the simplest of things.

就是一个ZIP文件。

It was like a ZIP file.

对吧？

Right?

是别人创建的。

Somebody else had created it.

但好吧。

But, okay.

所以我改进了我的起点。

So I've I've improved my starting point.

然后我们有了更强大的命令行界面、脚手架工具之类的。

And then we got stronger command line interfaces, scaffolding, that type of thing.

于是你就不必再担心具体的起点了。

And then you didn't have to worry about the exact starting point.

我觉得这只是让我们向前迈进了一小步。

And I feel like this is just taking us a few steps forward.

它让我们更容易启动一个基本可用的解决方案，但仍然要求你真正理解自己写入代码库并交付给用户的内容，归根结底，你不能推卸责任。

Like, it's making it easier for us to bootstrap a solution that that kind of works, but still requires you to really understand what you are putting into your code base shipping out to users and and not, you know, getting rid of your responsibilities at the end of the day.

我发现，至少在过去一两年里，对我而言，回归第一性原理总是能带来很大帮助。

I find that, at least for me in the last year or two, going back to first principles, as always, like, has helped a lot.

我最近与谷歌DeepMind团队更紧密地合作，研究Gemini及其编程能力，这让我重新认识到许多技术背后的运作机制。

I've been, working a little bit more closely, with the Google DeepMind team, on Gemini and, its coding capabilities, and that has been a really good reminder for me of just understanding how a lot of these things work behind the scenes.

如果你还记得，训练数据到底是什么？

If you remember that, you know, what is what is training data?

它很可能是基于GitHub或开放网络上那些许可宽松的代码，而这些代码中的模式往往反映的是最低共同标准。

Well, it's very likely gonna be looking at permissively licensed, you know, code that's on GitHub or or out on the open web, the patterns in that code are probably gonna be reflective of, in many cases, like, lowest common denominator.

那些只是能运行的东西，它们会是最安全、性能最高、最易访问的吗？

Things that maybe just work, are they gonna be the most secure, the most high performance, the most accessible?

可能不会。

Possibly not.

因此，如果你记得训练数据本身仍需大量工作，才能让系统达到真正可用的生产质量水平。

And so if you remember that the training data itself, still requires a lot of work to, like, actually get things to a place where, you know, it maybe is of production quality.

当你使用大语言模型时，就会开始适当降低自己的期望值。

You sort of start start to set your expectations a little bit lower when when working with LLMs.

你意识到，是的，它很可能比我自己从别处复制粘贴代码片段做得更好，这很明显。

You realize, like, yeah, it will probably do a better job than me just copying and pasting snippets of code from somewhere else, obviously.

但我仍然很可能需要在此基础上进行一定程度的手动工作或仔细检查，才能获得最佳结果。

But I still very likely need to do a level of manual work or a level of diligence on top of this to get the best outputs.

是的。

Yeah.

这让我想起大约十年前，我们大量使用 Stack Overflow，因为当你搜索某个问题时，答案就在那里。

It reminds me a little bit of ten years ago or so, ten five, ten years ago, we used a lot of Stack Overflow because when you search for something, it was there.

比如，对于邮箱验证，你会搜索‘如何验证邮箱地址？’

And so, for example, for email validation, like, you would search, like, how do I validate an email address?

当时有一个 Stack Overflow 的问题，有二十个回答，其中一个评分最高，是一个正则表达式。

There was a Stack Overflow question, and it had, like, 20 answers, one of them was top rated, it was a regular expression.

而我做的，我想很多人也这么做，就是直接复制粘贴，因为我懒得去仔细定义邮箱的正则表达式，那比简单的东西复杂多了。

And what what I did, and I think what a lot of people did, is I just copied and pasted because I cannot be bothered to define exactly the regular expression of emails, which is a lot more complicated than just things.

所以我只是随便用了一下，如果只是用于不重要的地方，它确实能凑合用。

So I just kind of and if it was for nothing important, it it kinda worked.

但当时确实有人提出批评和警告，指出其中一些——虽然不是这个具体例子——存在安全问题或未考虑到边缘情况。

But there were then complaints and, like, flags raised that a a few of them, not this specific example, but they were insecure or they didn't account for edge cases.

很多开发者在心不在焉、觉得无所谓或者明明知道更好的做法时，还是会这么做。

And a lot of devs, again, when you were in the mindless mode or it didn't matter or you know better, you did that.

我想我们在更大规模上也会遇到同样的问题。

And I guess we're we're we're gonna have the same thing at larger scale.

比如，这个bug为什么会发生？你回头去看历史记录时就会发现。

Like, why did the bug bug happen, and you're gonna look back at the history.

是的。

Yeah.

有人提交了一个看起来不错的大型请求，结果发现它是AI生成的。

Someone just looks good to me for for that big poor request that was in the end generated by an AI.

我觉得这只是一个很小的点，属于附带讨论，但我确实见过不少情况，现在有人会说：嘿。

That I think that this is this is very much a smaller point, a side discussion point, but I I've seen a number of cases where people will now say, you know, hey.

如果我能自己用提示生成一个稍微小一点的解决方案，那我还需要使用第三方库吗？这让我立刻想起了当年复制粘贴Stack Overflow热门回答的情形。

Do I do I still need to use third party libraries if I can just prompt a very slightly smaller version of a solution myself, which reminded me exactly of the Stack Overflow copy pasting the top response thing.

如果你是一名资深工程师，你会意识到：等等。

And what you re if you were a senior engineer, what you realized is, well, hold on.

这也意味着你现在需要承担起确保这些代码在未来对安全问题、各种平台问题都具备鲁棒性的责任。

That also means that you are now taking on the responsibility of making sure that this is future proof for security issues, for all sorts of platform issues that could happen.

如果你依赖某个库，那么这些修复可能更容易在一个集中点进行，然后部署给所有人。

If you're relying on a library, it's perhaps easier for there to be a central point of leverage where those fixes can be made and then deployed out to people.

如果你自己在代码库中维护所有这些不同的模式，那就意味着你也承担了这份责任。

If you own all of these different patterns in your code base yourself, that also means that you're taking on that responsibility.

这完全可以接受，但我有时看到人们并没有意识到，这需要额外付出一点精力。

Can be totally okay, but it's something that I I sometimes see people just not being mindful of of requiring a little bit of extra

我觉得，这时候我们只需要提醒自己，这不过是工程问题而已。

And I I feel this is where we just need to remind ourselves that it's just engineering.

对吧？

Right?

这是一种权衡。

It's trade off.

你是否愿意承担维护这个东西的责任和风险，比如它可能存在缺失的边界情况，或者无法正常或安全运行？

Do you take on the responsibility and the risk of of maintaining this thing and the fact that it might have missing edge cases or whatever or might not work correctly or securely?

还是选择依赖一个本身也有问题的第三方库？

Or do take on a dependency, which has its own problems.

对吧？

Right?

现在确实如此。

There's now Yeah.

依赖项的安全风险等等。

Dependency security risks, etcetera.

但说到软件工程的需求，有哪些新的工作流程，是一些我们以前作为软件工程师无法做到、但现在可以借助这些AI工具进行尝试的全新方法？

But speaking of the need for software engineering, what are some new workflows, some some things that we have not been able to do before as software engineers that we can now experiment with, that you you might be trying out with with these AI tools that is is just like is brand new?

是的。

Yeah.

对我来说，最让我期待看到发展的，是异步后台编码代理这一概念。

I think for me, the the thing that I'm most excited about seeing evolve is this idea of asynchronous background coding agents.

目前有几个团队正在探索这些想法，比如 Jules、codecs，还有 codecs。

A number of teams playing around with these ideas at the moment, Jules, codecs, the codecs.

我们还看到 GitHub 也在尝试一些类似的想法。

We also see, you know, GitHub experimenting with some of these ideas as well.

我认为，能够将部分待办事项委托出去，并让系统异步地完成这些任务，是一个非常有趣的想法。

I think that the idea of you being able to delegate parts of your backlog and have a system be able to implement that in you know, asynchronously is is a very interesting idea.

如果我们能以一种易于人类验证的方式避免合并冲突，再次回到‘人在环中’的理念，我觉得这非常有意思。

If we can do that without merge conflicts in a way that is easily human verifiable so, again, going back to that human in the loop, I think that's very interesting.

我发现，现在这种情况已经相当成熟了：如果你让一个智能体负责编写或更新你的测试，或者多个智能体协同工作，试图将你的代码库从某个库的版本迁移到另一个版本，或者从某个依赖项的版本升级到另一个版本，它们现在在这方面已经相当不错了。

I have been finding that that is very, very, very much already in a place where if you are having one agent work on writing or updating your tests, or you have a number of agents working together on trying to migrate your code base from one version of a library to another or, you know, a version of a dependency to another, that they're pretty okay at doing that kind of work right now.

对于更小的改动，比如我们每个人都要做的添加深色模式这类任务，如果能将这类小改动委托给异步智能体，未来它们会变得非常出色。

Smaller changes, like, there there are lots of things that, you know, we we all have to do, like adding dark mode, for for example, like smaller kinds of changes where maybe being able to delegate those kinds of changes to asynchronous agents are are gonna, you know, get get very, very good.

我对这一点感到兴奋。

I'm excited about that.

我认为，关于作为这场交响乐的指挥者，究竟什么样的交互界面才是合适的，以及你同时管理多少项任务才是现实的，这些问题的答案还有待观察。

I think that the jury's still out on exactly what is the surface for being able to you know, if if you are the conductor of this orchestra, what is the right surface for you being able to manage all of these things, and what is a realistic number of tasks for you to be managing at the same time?

因为，你知道，就连我，还有不少人都展示过，嘿。

Because, you know, e even I, like a number of people have shown off, like, hey.

是的。

Yeah.

我开了二十多个终端，其中一半在运行Claude代码，另一半在运行Gemini，看起来挺搞笑的，但事实上你的注意力是有限的。

I've got, like, 20 different terminals open, and I have Claude code running in half of them, and then Gemini running in and and it's it looks it's funny, but the reality is that you only have a finite amount of attention.

如果你真的想在代码审查和每个工作流程中投入精力，你可能只能同时处理几件事。

And if you are going to be actually putting in diligence into your code reviews and into each of these workflows, you are probably only going to be able to do a couple of things at the same time.

你知道，这一切最终还是回到多任务处理的最佳实践。

You know, it all comes back to to multitasking best practices.

但我对这个很感兴趣。

But I'm interested in that.

我很想看看这会走向哪里。

I'm interested in seeing where that goes.

另一件非常有趣的事情是氛围设计。

Another thing that's very interesting is is vibe designing.

我看到产品工程和产品开发的这一部分正在发生一些变化。

I'm seeing that part of product, you know, product engineering, product development evolving a little bit.

看到Figma的MCP正朝着这个方向迈进，这令人兴奋，这些工具能让设计师和开发者更紧密地合作，或者至少让设计师能够将自己的构想转化为可运行的原型，一种更接近代码、可以投入生产、而不仅仅是一个一次性演示的东西？

It was very exciting to see, you know, Figma's MCP moving a little bit closer in this direction, things that allow designers and developers to work more closely together or to at least enable designers to be able to take their vision and turn it into a functional prototype, something perhaps a little bit closer to code that can then be productionized and and is not just a one off demo?

是的。

Yeah.

我听说Shopify的一位工程主管或设计负责人提到，她团队里的每一位设计师——不是Shopify所有设计师——都在使用Cursor，他们先创建一个Figma设计，然后让Cursor来实现它，再展示给工程师看。

I I I heard an engineering lead or design leader at Shopify said that her team, every designer on her team, so not all of Shopify, has Cursor, and what they do is they create a Figma design, and then they ask Cursor to implement it, and then they show it to engineers.

这并不是说直接上线，而是我们现在有了一个可以互动的东西，而不是过去那种仅仅分享Figma设计的做法。

And this is not saying ship it, It's just more we now have something interactive that we can work with as opposed to just what they used to do, which is sharing a Figma design.

这真是我以前没听过的做法。

And I was like, this is something I've not heard before.

就像是，是的。

Like Yeah.

真的。

Truly.

我的意思是，所有设计师，或者至少是使用开发者工具的团队里的设计师，我无法想象设计师会用 Visual Studio Code，我的意思是，我或许能想象他们能打开它，但根本不可能想象他们真的会用，因为这根本不是为他们设计的。

Like like, all designers or at least on a team using the developer tool, like, I couldn't imagine designers using Visual Studio Code for I mean, I can imagine them being able to open it, but I couldn't imagine them it's just not built for them.

所以听到这个真的非常有趣。

So this was really, really interesting to hear.

再说一遍，这并不是什么供应商的广告。

Again, this wasn't some vendor advertisement or anything.

是的。

Yeah.

向 Shopify 团队致敬。

Shout out to the Shopify team.

说实话，我觉得他们中有不少人非常愿意分享团队中行之有效的方法，我很喜欢关注他们的故事。

Honestly, feel like a number of them have been very open to sharing what's been working out well for their teams, and I've enjoyed following their story.

我认为，这些模式能否更广泛地推广，将部分取决于培训、治理，以及明确区分原型代码和生产代码的界限。

I think that, you know, more of these patterns spreading is gonna depend a little bit on on training, on governance, on, like, clear boundaries between what is prototype code in production.

但已经能够把静态设计转化为半功能原型，这已经非常酷了。

But already being able to turn something static into a semi functional prototype is very cool.

我们会看到每个人都使用Cursor吗？

Are we gonna see everybody using Cursor?

我不确定。

I I don't I don't know.

我认为至少对我来说，我会看到人们通过他们最常使用的工具，或者工具之间的桥梁不断改进，来实现相同的结果，但我仍然觉得这非常令人兴奋。

I I think that I at least for me, I I see folks being able to accomplish the same outcomes from the tools that they spend the most time in or the bridges between tools getting better over time, but I still think it's very, very exciting.

同样地，我们也看到很多关于产品经理和工程经理角色变化的精彩讨论。

In the same way, you know, we're also seeing lots of good conversations about PM and EM roles changing.

产品经理可能会花更多时间在问题定义、指标和代理策略上。

PMs maybe, you know, spending more time on problem framing and metrics and policies for agents.

工程经理则会花更多时间在评估、安全审查以及真正帮助团队自信地使用AI上。

EMs spending more time on evals and safety reviews and really enabling their teams to work with AI confidently.

这不会改变对结果的责任归属，但我看到很多关于产品工程中‘品味’重要性的良好讨论，而这将是区分人的关键。

It won't change accountability for outcomes, but I am seeing a lot of good discussions about the need for taste in product engineering where that's gonna be what differentiates people.

因为如果任何人都能看看你构建的东西，并通过提示实现类似的功能，那么‘品味’这一部分在未来将继续成为人们关注的重点。

Because if anybody can, you know, look at what you've built and can use prompts to accomplish, you know, a similar set of functionality, the taste piece is gonna continue to be very interesting for folks to focus on in the future.

你和我之前讨论过初级和高级工程师的问题。

You and I have talked about sort of juniors and seniors.

我觉得，对于新毕业生和资深工程师来说，未来会很有意思。

I think that, you know, there's gonna be interesting times for new grads, versus seniors.

你知道，AI确实提高了入门门槛，但也提升了上限。

You know, AI definitely raises the floor, but it also raises the ceiling too.

初级工程师能够更快地上手，但那些能撰写规范、分解任务、理解系统架构、有效评审的资深工程师，我认为他们会变得更加宝贵。

And, juniors are gonna be able to get moving faster, but, seniors who can, you know, write specs, decompose work, understand system architecture, review effectively, I think that they're gonna become even more valuable.

我们之前谈到的最后那30%，我觉得是杠杆效应。

And that last 30% that we talked about, I think it's leverage.

不是单纯的琐碎工作，而是真正的杠杆效应。

Not just busy work, but actually, I think it's leverage.

很多调查都显示，目前人们对信任的态度是谨慎而乐观的，但这种谨慎正说明了人类监督仍需保持核心地位。

And, you know, a lot of surveys have been showing that trust is in a a cautious but optimistic place at the moment, but that cautious piece speaks to the need for that human oversight remaining pretty central.

是的。

Yeah.

当我想到并行代理这个概念时，我刚刚在想，从开发者角度来看，这种事以前从未发生过。

And if I think about, I was just thinking that when it comes to this idea of parallel agent, on one end, I don't think it's ever been done in the sense that as a developer, you were never able to do this.

我的意思是，我们以前根本无法用自然语言启动对话，让机器直接生成能编译的代码。

I mean, we we couldn't even kick off and talk with natural language to a machine that would spit out code that actually compiles.

我觉得这也是全新的，但更重要的是，你能同时用多个代理来完成——这在编程领域以前从未出现过。当你处于工作流中，专注于一个问题时，一旦停下来，就会切换到另一个任务，你的上下文几乎就像栈被清空了一样，然后你加载新的内容。

Like, I think this is new as well, but the fact that you can do it with multiple, on one end, it's not happened before on programming is a very, you know, you're in the flow, you work on one problem, when you stop working on it, you switch over, you get your context almost like the stack clears, you know, you load the new thing.

这种感觉确实有点像。

It it kinda feels like it.

对吧？

Right?

但另一方面，当我回想起我认识的那些最优秀的资深工程师，他们的日常是怎样的呢？他们通常和几个中级工程师、可能还有一些实习生或新毕业生组成团队，各自处理自己的任务。

But on the other, when I think back to who are the best senior engineers I know and what do they their days look like, well, they're on a team with a few mid level, may maybe some interns or new grads, and they're working on their stuff.

然后，他们往往是那个收到Slack消息的人，消息写着：嘿。

And then they're the ones who get the Slack ping saying, hey.

你能帮我解除阻塞吗？

Could you please unblock me?

所以他们会切换上下文，审查代码，批评它，或者专门安排时间反复审查。

So they context switch, review the code, you know, like like, criticize it, or they block out time and they're review, review, review.

而且，对于每一位资深工程师——通常他们也是技术负责人——都会带几个下属，他们不是代理，而是真人，但他们会审查他们的工作，并在站会上、计划会议上稍微引导他们，给予提醒和指导。

And, like, for every kind of senior slash often these are tech leads, they have, like, a few people who are you know, they're not agents, they're people, but they they just review their work and they kind of orchestrate them a little bit on the stand up, you know, they're in the planning meeting, they nudge them, they mentor them.

所以某种程度上，我觉得我们资深工程师已经在做这件事了。如果我有一根魔法棒，问：几年后谁最有可能管理多个代理？

So in some sense, I feel like we seniors already do this, and if if I had a magic wand saying, like, who's who would I expect in a few years' time to be able to manage multiple agents?

肯定是资深工程师。

Well, the senior for sure.

至于新人，我觉得对他们来说这可能太勉强了，而且他们本来也不被期望具备这种能力，因为他们缺乏经验。

Like, new grads, probably, like, I mean, I I feel that will be a stretch for them, but they're not expected to have that, and they don't have the expertise.

所以我很好奇，这些技能是否在某种程度上是可迁移的？为什么这位资深工程师能做得这么好？

So I I wonder if some of these skills are somewhat transferable in the sense that and, you know, that senior, why could that senior do that well?

因为他理解了代码库。

Because he understood the code base.

他知道什么是好的代码。

They knew what good looked like.

他们总是非常细致地进行代码审查。

They were always thorough on their reviews.

他们从不让任何问题溜走。

They never let things slip.

他们会指出哪怕是最微小的细节。

They called out even the smallest thing.

我完全同意你的观点。

I completely agree with you.

而且我认为，团队中的开发者教育也可能为此发生演变。

And I and I think that there's a lot to be said about how developer education in Teams may also evolve for this moment.

回想我刚入行的时候，导师制一直是个重要话题，我们经常讨论，特别是对于刚加入新团队的人，结对编程的重要性。

Historically, I I remember when I was coming up, you know, mentorship was always a big topic, and we talked about, you know, especially for folks, who are joining a new team, the importance of, like, pair programming.

我想我们会看到类似‘三人编程’的情况，即一位初级开发者、一位资深开发者和AI共同协作。

I think we're gonna see, you know, perhaps even situations of of trio programming where it's a junior senior and the AI.

也许资深开发者会要求你解释AI生成的代码，或者引导你理解这段代码如何与系统的其他部分关联。

And maybe the senior is going to be there asking you to explain the code that the AI is generating in some way or walking you through how that code perhaps connects to other parts of the system.

而且，再次强调，把它当作工具箱中的额外工具，帮助提升对整个应用程序如何运作的信心和认知。

And, really, again, using it as an additional tool, in their arsenal, to be able to help, build up confidence and awareness, of how of how the full app actually works.

我们还看到一些关于潜在新角色或角色优化的有趣讨论，比如前线工程师。

We're also seeing some interesting discussion about, you know, potentially, new roles or or refinements of roles, things like forward deployed engineers.

我注意到人们对那些更贴近客户的开发者感兴趣，他们能利用AI快速开发功能，同时反馈需求。

I'm seeing interest in, you know, developers who are a little bit more embedded with customers, who can build features rapidly with AI while feeding back requirements.

这类情况可能会模糊开发者、产品经理和设计师角色之间的界限。

And that type of thing might blur the lines between, you know, the developer, PM, and designer roles a little bit.

我很想知道这些趋势最终会走向何方。

I'm interested to see where where that, may end up going.

我也很感兴趣，从广义上讲，AI工程将如何改变我们对教育的思维方式。

And I'm also interested in how, you know, just generally speaking, AI engineering is going to evolve how, you know, we approach education.

无论你是在高中还是大学，我们会不会教人们提示词和上下文工程的最佳实践？

Whether you're in high school or you're in college, like, are are we gonna be teaching people prompt and context engineering best practices?

这一切到底会是什么样子？

What what does that all actually look like?

我们如何继续帮助人们以系统设计和工程的思维来思考？

How do we continue to enable people to think with systems design and engineering in mind?

但我对教育这一方面的发展非常期待。

But I'm I'm very excited where the education side of this goes.

我们之前讨论过的一个领域是代码审查，它真的非常重要。

So one area that we've talked about was code reviews and how it's really important.

这是一个瓶颈，我们应该审查代码。

It's it's a bottleneck, and and we should review the code.

我已经注意到自己的一种倾向：当我使用这些AI工具时，无论是ClotCode、代理，还是自动补全功能，我都会习惯性地点击、接受，甚至直接接受所有建议，尤其是在处理那些并非至关重要的任务时，因为这些任务看起来很简单。

One thing that I'm kind of noticing on myself already is when I use these AI tools, may that be ClotCode or or agents or or even just the autocomplete, I have a tendency to do tap tap or accept or just, like, accept all the things, especially when I'm working on something that, I mean, it's it's not the most critical thing in the world, and it's kind of it's easy.

尤其是当我开始信任这些工具大多数时候都能正确完成任务后，我知道最终我还是要审查的。

And especially after I started to trust that it gets it right most of the time, in the end, I know I'm gonna review.

但我发现自己在批判性思考上有所减弱。

But I find myself losing a little bit of criticality.

我没有以前那么挑剔了，比如第二天、第三天，当我刚接触这些工具、还不信任它们的时候，我的态度可没这么松懈。

I'm not as critical, you know, the second day and the third day and and when I was the first time when I didn't trust this thing.

我有点担心，在代码审查中，同样的情况也可能发生。

And I worry a little bit that with code reviews, the same could happen.

你知道，LGTM，看起来没问题。

You know, LGTM looks good to me.

我明白，这是谷歌的一个隐患，而且被内置在你们的代码审查工具中，据我了解，这是一个非常有趣的工具。

I mean, I understand that's a Google hazard and and built in as a as a as a feature in in your code review tool, which is a really fun tool, as I understand.

但确实存在这种风险。

But there is this risk.

当然，一直以来都存在这种风险，就是你只是觉得工作量太大，

There's always been this risk, of course, that that, like, you just kind of, like, it's a lot of work.

看起来没问题，但你并没有认真审视。

It kinda looks good, and you're not looking critically.

你觉得，尤其是在正规的团队中，我们该如何应对或对抗这种敷衍了事的审查行为，特别是当我们自己写的代码并不多的时候？

How do you think, especially on on on, you know, like, proper teams, we could battle or or or we could do this kind of fighting against, like, yes, giving it a proper review, especially if we haven't written the code that much?

因为我觉得，写代码的时候，你至少会有两次审查。

Because I feel with writing the code, you have, like, two reviews.

一是你在写自己的代码。

One is you're writing your own code.

你亲手敲出代码，虽然我们现在不常这么做，然后别人在知道这是你写的代码的情况下进行审查。

You're typing it out, which we're not doing as much anymore, and then someone else gives a review knowing that it was you who typed it out.

我的意思是，写代码总是比阅读和审查代码更有趣。

I mean, it's always it's always more fun to to write code than to to read and and review it.

是的。

Yeah.

有一点。

A little bit.

对吧？

Right?

对。

Yeah.

我认为我们工作中的越来越多部分将变成阅读和审查代码。

And I I I think more and more of our job is going to become about reading and reviewing code.

我曾尝试向大家提出一些想法。

There are a few ideas that I've I've tried to suggest to folks.

比如，对于某些功能或特定的星期几，你可以有意地尝试不依赖AI或大语言模型，只是看看：好吧。

Things like, you know, for certain features or certain days of the week, maybe you intentionally try to not lean on using AI or an LLM and just try to see like, okay.

我还能不能自己解决一些这些问题，以保持你的批判性思维能力，并迫使你思考：好吧？

Well, can I can I still solve some of these problems myself to preserve your critical thinking skills and force you to think about, okay?

假设所有的顶级大语言模型提供商一整天都宕机了，你会怎么做？

Well, let's say that all of the top LLM providers were down for the day.

你会怎么办？

What would you do?

开个玩笑，我想我会说，是的。

Being cheeky, you know, I would probably say, like, yeah.

我就直接用Ollama和本地模型，完全没问题。

I'm just I'm just gonna go, you know, use Olama and a local model, and I'll be totally fine.

我会找到一些备用方案。

I will find some fallback.

但现实是，我认为对于我们来说，能够独立思考事物的工作原理、不依赖AI就能解决问题，仍然会非常重要。

But the reality is that I do think that it's gonna continue to be very important for us to be able to think through, how things work, be able to problem solve without necessarily, you know, relying on the AI.

这些模型会继续变得越来越好。

The the models are going to continue getting better.

它们从你的代码库中获取足够上下文的能力也会持续提升。

Their ability to take in sufficient context from your code base is going to continue getting better.

但在我看来，在你能完全信任AI在任何情况下都能正确应对你提出的各种需求之前，还需要相当长的时间。

But it is going to take quite some time, in my opinion, before you're going to be able to fully trust that in every situation, whatever requirements you throw at it, it's gonna be able to get it right.

如果你卡住了，尝试了五次、十次还是没能解决问题，你就得自己来解决这个问题。

And if you get stuck, if you're trying things out, and you've tried it five, ten times, and it still hasn't solved the problem, you're gonna have to solve the problem yourself.

所以我认为，主动创造一些情境来锻炼和反复测试自己的批判性思维能力，将会非常重要。

So I think that, being able to force yourself into situations where you're testing and retesting your critical thinking skills are gonna be important.

我还觉得，在这方面做一些博弈论式的思考会有些价值。

I also think that there's a little bit of value in, you know, doing some game theory around this.

团队或个人可能会越来越依赖AI来完成更多的代码审查，以跟上快速变化的节奏。

Teams or individual people are probably going to start leaning on AI to do more of the code reviews to keep up with the, you know, pace and and velocity of change coming in.

这些工作流程会是什么样子？

What are those workflows going to look like?

如果你有一个代理说，是的。

If you have an agent that's saying, like, yeah.

我审查了这个PR，看起来可以合并了。

I reviewed this PR, and it looks good to merge in.

你真的会信任它吗？还是会亲自进行人工审查，即使你的时间投入可能比以往更少？

Are you actually gonna trust it, or are you gonna go and do a human level review even if it take you know, is maybe more shallow than you historically would have done?

或者你愿意花上十到二十分钟来完成审查。

Or maybe you're okay with spending ten, twenty minutes or something on on the review.

那么，这种工作方式会是什么样子？

Like, what does that work will look like?

就我而言，即使一个代理告诉我某件事可以合并，这也是一种信号。

In my case, even if an agent an agent telling me that something looks good to merge is a signal.

这类似于一种信号，比如：好吧。

It's similar to a signal of like, okay.

也许我团队里的一名初级成员也给出了 LGTM。

Well, maybe a a junior person on my team has also, you know, given this in LGTM.

但如果足够关键，我可能还是会回去检查。

I'm still gonna probably go back if it's critical enough.

或者 CI/CD 通过了所有测试。

Or the CICD passes all the tests.

对吧？

Right?

状态是绿色的。

It's it's green.

所有测试都通过了。

All tests pass.

没错。

Exactly.

没错。

Exactly.

这只是一个信号。

It's it's a signal.

信号很有用，但你仍然需要运用自己的判断标准，仔细检查，确保自己有信心。

Signals are useful, but you still want to apply, you know, your lens on quality to it and and take a look through and just make sure that you're you're confident.

像拥有这些测试、你和你的团队讨论过的各种质量检查点，把这些都落实到位，这些都是建立信心的好信号。

Things like having those tests, having, you know, whatever quality gates that you you and your team discuss, like, having those in place, all of these are good signals for building up confidence.

因此，你拥有的信号越多，就越能增强信心，确保合并代码时不会出问题，我认为这是好的。

So the more signals that you have can build up to build up confidence that things aren't gonna go off the rails when you merge in, I think that's good.

但我尽量有意识地确保自己不会完全依赖AI来做所有事情。

But I try to, you know, be very intentional with making sure that I'm not leaning on AI for absolutely everything.

我可能会在日常生活的许多方面使用AI，包括AI编程，但我仍然会确保，无论有多少任务——比如20%、30%或任何比例——并不一定需要AI，我都会尽量用自己的大脑去解决这些问题。

I may use it for many aspects of my life on a day to day, including AI coding, but I still try to make sure that I'm you know, whether it's 20 or 30% or whatever amount of my tasks don't necessarily require AI, I still try to make sure that I'm using my brain, to to solve those problems myself.

我认为这种主动性——主动保持批判性思维能力——会对人们有所帮助。

And I think that intentionality, like, just be proactive with maintaining your critical thinking skills, I think that's gonna help people.

是的。

Yeah.

关于这些事情中的一点，你仍然要亲力亲为，我想到两个例子。

And one of these things of like, you're still hands on two things came to my mind.

其中一个就是最近上线的 Cloud Code。

One was a Cloud Code actually recently shipped.

你可以切换模式。

You can change modes.

你可以开启‘解释模式’，它会详细解释内容。

You can have Explosatory Mode where it explains things.

还有一个叫‘学习模式’，它会暂停并让你自己完成某部分，我觉得这非常巧妙。

And you have a thing called Learning Mode, which is it pauses and says, you do this part, which I thought was really clever.

我确实为我正在开发的一个项目开启了这个功能。

I actually turned it on for for something that I was building.

它没有像我预期的那样工作，因为它给了我一个非常奇怪的任务，但我看到了它的潜力。

It didn't work as I expected because it gave me a really weird task to do, but I see the potential.

我其实有点想更多地依赖它，也希望其他工具也能将这种功能纳入开发者工具中。

And I actually I actually am tempted to lean a bit more on that, and I think it's I hope other tools will also do this as part of the developer.

我觉得我确实想知道我能不能做到。

Like, I I feel I do want to know that I can do it.

我能做，但如果不做，我怎么知道我是不是做对了呢？

I I can do it, but how am I gonna know if I'm if I'm not doing it?

当然，你会稍微摆弄一下

Of course, you fiddle around a little bit with

我觉得这个想法太棒了。

I think that is such a fantastic idea.

如果你是新人，或者刚接触一个不熟悉的代码库，别急着去尝试提供价值，直接催生一个新功能。

And if somebody is a junior or you are coming into a code base that you're not familiar with, fight that, feeling of the first thing I'm gonna do is try to provide value, and I'm gonna just prompt a new feature into existence.

不如先用大语言模型来解释一下这个代码库是怎么运作的，花点时间沉浸其中，好好理解这个代码库的丰富内涵。

Maybe use the LLM to just explain how the codebase works, and just spend time, like, soaking yourself up in the richness of, like, what is the codebase?

它是怎么工作的？

How does it work?

所有部分是如何相互连接的？在开始催生功能之前，先搞清楚这些。

How does everything connect together before you start prompting things into existence?

我认为把它用作学习工具是非常强大的。

I think that using it as a learning tool is is a very powerful thing.

我们需要更多的这种做法。

We need we need more of.

你提到的这一点，新员工使用学习工具，这也是很重要的一点。

And this is also something, what you said, new new new joiners using a learning tool.

我听到一些公司说，新员工的入职速度变快了。

I'm hearing from a few companies that they're seeing new joiners onboard faster.

那些能够使用这些AI工具来解释内容、梳理代码库的新人，把这些工具当作一个值得信赖的人。

The ones where they can use these AI tools to explain stuff to them, outline the cloud base is is just a a kind of a trusted person.

你可以24小时随时提问，就像问一位资深工程师一样，我的意思是，你随时都能问，但显然你并不想频繁打扰。

Twenty four seven, you can ask questions on, like, a senior senior engineer who I mean, you you can ask questions anytime, but, obviously, you don't wanna do it.

毕竟，一天只有八小时工作时间。

It's, like, you know, eight hours of the eight hour working day.

你会给自己设定界限的。

Like, you you're you're gonna keep your limits.

有趣的是，这种变化改变了许多使用这些工具的工作场所的氛围。

And it's it's interesting how that that has changed the dynamics at the at a bunch of workplaces that do have these tools.

确实如此。

Absolutely.

我认为，我希望将AI作为学习工具的做法能成为一种更普遍的标准。

I I I think that, you know, my my hope is that it's gonna become a much more standard thing, treating AI as a learning tool.

这不仅仅适用于理解新的代码库。

And it's it's not just for understanding, you know, a new code base.

它对于理解编程概念、框架或架构模式也非常有用。

It can be very useful for understanding programming concepts or frameworks or architectural patterns.

有时我想把一个功能从一个截然不同或使用不同编程语言的代码库迁移到另一个代码库。

There are times when I wanna be able to bring a feature over from one code base that is very different or written in a different programming language over to another.

我发现，在这些情况下，AI对于帮助我理解一个代码库并找到将其内容迁移到新代码库的路径是不可或缺的。

And I have found AI, very much indispensable for those situations of helping me understand one code base and the path to being able to bring over something, to to a new one.

因此，鼓励你的团队尝试AI工具，分享最佳实践，让使用AI作为学习工具成为常态，我认为这对你作为领导者和管理者来说将大有裨益。

And so encouraging your team to experiment with AI tools, sharing their best practices, making it a regular thing that people know it's okay to use it as a learning tool, I I think it's just gonna be great for you as as as leads and and and managers.

你也是一个团队的管理者。

And you're you're also a manager of of a team.

作为其中的一部分，你当然要进行绩效评估和晋升，帮助人们在职业上更好地成长。

And as part of this, obviously, you you do performance reviews, promotions, but help people basically professionally get better.

自从这些工具出现以来的过去三到四年里，你对优秀且扎实的软件工程师的定义，或者你在团队中看到的定义，发生了怎样的变化？

How do you think the kind of your definition or what you're seeing on the team, a definition of a really standout solid software engineer has changed in the last three to four years since these tools have come out?

有什么新的变化？又有什么是保持不变的？

What is new and and what is the same?

我认为新的变化在于终身学习的重要性——我始终认为这一点至关重要。

I think that what is new is the importance you know, we one of the things that I have found has always held true is the importance of being a lifelong learner.

这一点一直都没有改变。

That has always remained true true.

无论框架如何更替、工具如何变化，行业如何发展。

Doesn't matter if frameworks come and go, tools change, you know, the industry evolves.

保持终身学习的态度，乐于尝试新事物，敢于失败，并不断积累技能，我认为这依然非常重要。

Being a lifelong learner and being very open to trying out new things, failing, building up those skills, I think that that remains very, very important.

我团队中那些早期就能成功利用AI进行编码和产品工程的人，都是带着这种心态的：我乐于学习新事物，愿意尝试，并拥有成长型思维。

The people on my team that I think were the most successful early on at leveraging AI for coding and for product engineering were the ones that went in with this mindset of, I'm happy to learn new things to try out and to have this growth mindset.

如果没成功，也没关系。

And if it doesn't work, that's fine.

至少我尝试过了。

At least I've tried it out.

至少我理解了其中的限制，未来也许会为不同的使用场景尝试不同的模型。

At least I understand, the constraints, and maybe I'll try out different models for these different use cases in the future.

我认为这一直是一个非常重要且一贯的特质。

I think that that has been very much a consistent, thing that's been important.

我觉得，作为团队负责人，现在正是你帮助团队度过这一阶段的时候，要展现出你自身也愿意学习。

I feel like if you are a lead, now is the time for you to be helping your team through this moment in terms of showing that you're okay with learning as well.

我每周都会做的一件事是，花大量时间阅读，我非常喜欢阅读。

One of the things that I do every week, actually, is I spend a lot of time I love reading.

我花很多时间阅读。

I spend a lot of time reading.

我当然会经常阅读你的通讯。

I will, of course, read your newsletter a lot.

我会阅读大量不同的论文、白皮书、博客，观看视频和课程，关注每周AI和AI工程领域的新动态，并将这些内容整理成通讯发给我的团队。

I will read a number of different papers, white papers, blogs, watch videos, watch courses, read announcements about what is new in AI, what is new in AI engineering during the week, and I will surface those things to my team in a newsletter.

我也会为你的团队制作一份内部通讯吗？

I I will also surface those an internal newsletter for your team?

我有一份内部通讯。

I've got an internal newsletter.

我每周一都会写它。

I I write it every Monday.

所以我会在本次对话结束后继续写它。

So I'll be I'll be writing it after after this.

我会包含一些我正在尝试的项目。

And I will include, like, what am I hacking on?

我正在写些什么？

What am I writing?

我在想什么？

What am I thinking?

我认为我们的团队真正需要注意哪些事情？

What are the things that I think are important for our team to actually pay attention to?

在这个时代，你知道，你看到这么多人难以跟上进度，如果你在推特或其他社交媒体上关注动态，可能会感觉非常

And in this time where, you know, you you see so many people struggling to stay up to date, if you follow along on on Twitter or other social networks, it can feel very

令人不知所措。

overwhelming.

是的。

Yeah.

因为每隔几个小时，就感觉有什么东西从根本上发生了变化。

Because every few hours, it feels like something has changed in a fundamental way.

对。

Yeah.

能够筛选这些信息，帮助人们专注于真正重要的事情，我认为这正是当前领导层应该做的，特别是如果你能引导大家关注，嘿。

And being able to sift through that and help people to just pay attention to, like, what is actually important, I think that's a great thing for leadership to be doing right now, especially if you can guide folks towards, hey.

也许可以多花点时间研究这个，而不是那个。

Maybe spend a little bit more time poking at this rather than that.

我只是想说，我觉得这很符合伊斯兰教的风格，因为你仍然更贴近行业本身。

I just want to say, I think it's Islam dunk because you're also staying closer to, you know, the the industry.

比如从技术层面来说，说你亲力亲为可能有点夸张，但通过持续关注这些，你已经非常接近了。

Like, to being technical, maybe it's a stretch to say hands on, but you're pretty close to that just by keeping up with all of this.

是的。

Yeah.

对。

Yeah.

而且我知道，我会让团队里的其他人来发表意见，但我总体上收到的反馈都很好。

And I've you you know, I I I'll I'll defer to, you know, people on my my team to to to say things, but I think that would I've generally had good feedback about this.

我甚至收到过其他高管说：嘿。

I've even had, like, other execs saying, hey.

实际上，我发现很难跟上这些进展，而你的通讯稿也在帮助我度过这个阶段。

Actually, I'm finding it really hard to stay on top of this, and your newsletter is helping me navigate this moment as well.

所以我认为，作为团队负责人，能够以你感到舒适或时间允许的任何程度跟上当前的发展，这对你的团队来说会是非常有力的，也能帮助你保持自己的技能相对敏锐。

And so I I think that as a lead, being able to stay on top of what is happening at whatever level you are comfortable with or that your bandwidth allows, I think that that can be a really powerful thing for your team and will help you just keep your your own skills relatively sharp.

我们之前聊了很多关于AI辅助工程的话题。

You know, we've been talking a lot about AI assisted engineering.

如果你们团队需要构建AI功能，很多这类内容对他们来说也会继续相关，因为你能够帮助他们厘清：在AI领域，哪些行业动态真正与他们的工作相关，哪些并不重要。

If your team do have to build out, like, AI features, a lot of this stuff is also gonna continue to be relevant, for them too because you can help connect the dots between what is happening in the industry where AI is concerned that is actually important to their work versus what is not.

尤其是在当下，这尤其重要，因为可能有些事情是他们过去从未需要考虑过的。

And that's especially I think it's exceptionally important in this moment where maybe there are things that they have not had to historically think about.

比如，嘿。

Like, oh, hey.

模型X在图像生成方面非常出色，但模型Y在生成声音方面更厉害。

Well, model x is really good at image generation, but model y is really good at generating sound.

我这只是随便举例。

I'm I'm just making things up.

但我认为，你能以更具体的方式指出这些机会，会非常有用，这样你的团队就可以自行深入研究，而不会从零开始。

But I think that it it is really useful for you to be able to, maybe highlight some of these opportunities in a more concrete way, and then your team can go off and and spend time digging into it themselves, but they're not starting from a blank slate.

而且，我想你只是在展示，看吧。

Plus, I guess you're you're just showing that, look.

看吧。

Look.

我在学习。

I'm learning.

我花时间阅读、理解和消化。

I'm spending time reading, understanding, digesting.

我在业余时间做些事情，这相当于表明这样做是可以的。

I'm doing some stuff on the side, which kind of gives the permission that it's okay to do it.

而且，说实话，每个人情况不同，但我想，鉴于我们现在正经历这项技术的重大变革，它是一个新工具，拥有许多新功能。

And if anything, I mean, you know, everyone for themselves, but I would assume that right now, because we are having a big change in in this technology, it's it's a new tool, a lot of capabilities.

在大多数公司和团队中，除非你正处于周五紧急交付的高压状态，否则让工程师花点时间实验、探索一下‘我能不能用这个东西？’应该是可以的。

It should be okay at most companies, and most teams, unless you're like has done crunch time shipping something like on Friday to for for engineers to spend some time experimenting, figuring out, hey, can I use this thing?

它对我有用吗？

Will it work for me?

你知道吗，当我再次与Shopify的同事们交流时，他们经常这么做，最终，有些人可能以为会发生的情况是：人们会做更少的工作。

You know, when I talk with, again, the the folks at Shopify, they do a lot of this, and and in the end, actually, what what some people think might happen is, like, well, people do less work.

但实际上，发生的是他们做的工作量和以前一样，只是更有动力了。

Actually, what happens is they do the same work as they did before, but they're way more motivated.

他们尝试了很多新东西。

They tried a bunch of new stuff.

有些尝试没成功，你可能会觉得这是浪费，但其实并不是。

Some of it doesn't work, and it's kind of a a would think it's a waste, but it's actually not.

这是一种学习过程，现在他们对哪些方法有效、哪些无效有了更大的信心。

It was learning, and now they're a lot more confident about what works, what doesn't.

你知道，他们会很自信地说：好吧。

You know, they'll they'll confidently say, like, okay.

我们不会用AI来生成代码库的这一部分，但用于原型设计是很好的，等等。

We're we're not gonna use AI for to generate this part of the code base for prototyping is great and so on.

所以我觉得，是的，每个人都在摸索，因此不获得许可几乎是种浪费，而有什么比你自己也这么做更能获得许可呢？

So I feel like, yeah, everyone's figuring things out, so it's it's almost a waste to not get permission, and how how how how better way to get permission than to show that you're doing this as well?

是的。

Yes.

是的。

Yes.

当然。

Absolutely.

人工智能是一种工具。

AI is a tool.

我认为，能够压缩工作量的工具，也可能带来清晰的时刻，而且尝试某些类型的想法比以往任何时候都更快。

Tools that can compress work, I think, can can also lead to good moments of clarity, and it is faster than ever to try out certain classes of ideas.

我认为，这种效率有时也能凸显我们人类品质的重要性，比如在判断我们应该推进什么时。

And I think that that efficiency can also sometimes highlight the importance of our human qualities when it comes to, like, judging What what should we be moving forward with?

我们应该忽略什么？

What should we be ignoring?

我们应该最小化什么？

What should we be minimizing?

我认为，在当下尝试拥抱人工智能，弄清楚什么对你和你的团队最有效，是充分利用时间的好方法。

I think that trying to embrace AI in this moment and figure out what, you know, what works well for you you and your teams is is a great use of time.

我想说，如果您的观众中有企业界的人士，尤其是我知道，经历过这段旅程后，可能会有一些团队觉得，哎呀，感觉初创公司比我们行动快得多，或者他们能使用所有这些优秀的新工具和模型，而我们却还在等待企业友好版的这些工具。

I did wanna say, if there are folks in your audience who are in enterprise, especially, I know that having gone through this journey, there may be teams who feel like, oh, man, it feels like startups are able to move, you know, so much faster than we are, or they're able to use all of these great new tools and these models, but we are still waiting on the enterprise friendly version of these things.

我曾与一些团队交流过，他们有时会觉得，我们在等待。

And I've talked to teams where they can sometimes feel like, we're we're waiting.

我们在等待。

We're waiting.

我们在等待能够尝试更多这些工具。

We're waiting to be able to try out more of these tools.

我在我团队里最终做的，是几年前我们也同样在等待公司官方认可的方式，但这并没有阻止我们学习。

What I ended up doing in my team was we were similarly you know, a couple of years ago now, we we were similarly waiting for the company embraced official way of doing things, but that didn't stop us from being able to learn.

所以我鼓励大家，我们很多人都在周末搞一些副项目，对吧？

So I encourage people, well, a lot of us are hacking on side projects, right, at the weekends.

我们可以尝试任何第三方工具、第三方模型，或者我们自己的模型。

We can try out whatever, you know, third party tools, third party models, our own models.

我们可以尝试和学习，这对你来说不一定要成为很大的障碍。

Like, we can try out and and learn, and that doesn't have to be a big blocker for you.

所以你仍然可以协助你的团队踏上这段旅程，而无需过多地等待。

So you can still, you know, help your team along this journey without necessarily having to to do a lot of that waiting.

所以只是想让大家知道。

So just wanted to let folks know.

通往拥抱这个时代的方式有很多。

There there are many paths to being able to embrace this moment.

是的。

Yeah.

我认为共识基本上是，这些工具将会广泛普及。

And I I think the consensus is pretty much that these tools will be widespread.

它们不断变化，而且已经对许多团队的代码库产生了贡献，人们正在使用它们。

They keep changing, and they are already contributing to a lot of teams' code base and people are using it.

所以不如直接使用它们，因为最终这需要时间。

So might as well just, you know, use it and and because it takes time in the end.

所以你越早开始，无论怎样都会有大量的学习要完成。

So the earlier you start, the you'll have plenty of learning to do regardless.

是的。

Yes.

完全正确。

Absolutely right.

作为总结，我有几个快速的问题，我会直接抛出来，然后你告诉我你第一反应是什么。

So as closing, I just have a few rapid questions, so I'll just fire them, and then you you tell me what what comes to mind.

你最喜欢的编程语言是什么？为什么？

What's your favorite programming language and why?

我最喜欢的编程语言是 JavaScript。

My favorite programming language is JavaScript.

这是一个非常有偏见的答案。

This is a very biased answer.

我本来以为你会这么说，但我不确定

I I thought you might say this, but I wasn't Of

当然。

course.

你写过关于My的书

You wrote books about My

我最喜欢的编程语言是JavaScript，这并不是出于个人原因。

favorite programming language is JavaScript, and it's it's not for the personal reasons.

可能还有更好的编程语言。

There are probably better programming languages.

我喜欢JavaScript，因为它让地球上任何人都能无需门槛地构建并发布网页内容。

I like JavaScript because it enables anybody on the planet to be able to build and ship something to the web in a way that doesn't require gatekeepers.

它非常开放，我认为这种理念非常自由。

It it's very open, and I think that there's something very liberating about that idea.

所以这可能是我最喜欢JavaScript的原因之一。

So that that is one of the reasons why I like JavaScript probably the most.

我喜欢它。

I I like it.

我总是很惊讶，当一个软件工程师说喜欢JavaScript的时候，因为我们都知道，与其他语言相比，它有很多限制，而且也已经有不少关于它的书籍出版了。

I I'm I'm always surprised when a software engineer says JavaScript because we know that compared to other languages, it it has a lot of limitations, and, you know, there's been books written about it as well.

但我无法不同意他的观点。

But I I cannot disagree with it.

我真的很喜欢它。

I I I love it.

这是事实。

It is true.

你最喜欢用什么工具？它有什么功能？

What's a tool that you really like using, and what does it do?

我觉得我目前最喜爱的工具之一，当然我有偏见。

I would say that one of my favorite tools at the moment, and I and I'm biased.

我也写过一本关于这个工具的书。

I have a I have a book out about this as well.

它可能是 boltbolt.new。

It's probably boltbolt.new.

Bolt.new 是一种用于快速搭建应用的 Vibe 编码脚手架工具，用户可以使用它。

Bolt dot new is one of those Vibe coding scaffolding tools that people can use.

他们最近刚刚增加了对自定义智能体的支持。

They very recently added support actually for being able to use custom agents.

因此，你可以使用 Claude Code 来构建你的 Vibe 编码应用，而且生成的结果通常质量很高，设计也很出色。

So you can use Claude Code, for example, to build your Vibe coded app, and, the the output is generally very high quality, great designs.

所以我一直很喜欢这个工具，而且这个团队在提供一些非常棒的集成方面，可以说走在了前沿。

So I've I've been really enjoying that, and and the team have been, I I would say, on on the edge of of offering some really great integrations.

我一直很欣赏这个想法：许多 Vibe 编码平台现在开始关注集成层的问题。

So I've been I've been liking this idea that many Vibe coding platforms are now starting to think about the integration layer.

那么，我们如何自动化地解决你需要使用像 Superbase 或身份验证提供商这类工具的需求呢？

So how can we automate the idea of you needing to use, you know, like, a super base or an authentication provider or these other things?

当然，你仍然需要留意生成的所有代码，但能减少这些初始设置的障碍，我觉得非常棒。

You still need to pay, of course, attention to all of the code that's being generated, but being able to remove more of that setup friction, I think, is amazing.

关于 Bolt，还有一个有趣的小知识。

And I guess the fun fact on Bold.

New 这个项目最初是一家名为 Stackblitz 的公司开发的，他们打造了一个非常棒的在线编辑器，功能先进，后来他们转向了 bold.new。

New is they started off as a company called Stackblitz, who built a really cool online editor, really advanced, and then they moved over from that to bold.new.

而且，背后有一群真正精通技术的软件工程师。

So and there are software engineers behind it, again, who really know their craft.

是的。

Yes.

是的。

Yes.

当然。

Absolutely.

最后，你推荐哪一本书？

And finally, what's a book that you would recommend?

我推荐的这本书，我认为它以一种让我非常信服的方式，涵盖了软件工程师的职业发展路径和最佳实践。

So the book I'm gonna recommend is one that I think covers, I would say, career paths and best practices for software engineers in a in a way that I found very compelling.

这本书是你写的。

It is by you.

这是《软件工程师指南》。

It's the software engineer's guidebook.

当然，我这是在迎合观众，但这本书确实非常出色。

And, of course, I'm I'm playing to the crowd, but it it is generally gen genuinely a very, very sharp book.

根据我读过的所有评论，其他人也认为这是一本极佳的读物。

From from all the reviews that I've read about it, other people have also found it to be an excellent read.

它就放在我桌上，不过我们刚才没提到这个。

It's it's on my desk, but we we didn't talk about this, by the way.

所以这个推荐对我来说就像对其他人一样，完全是意外。

So this recommendation came just as a surprise to me as as others.

谢谢。

Thank you.

这确实是一本非常出色的书。

Genuinely an an excellent an excellent book.

如果我不推荐这本书的话，我想如果你现在想更多了解人工智能工程的基础知识，有一本叫《AI工程》的书，作者是黄智，也值得一看。

I guess if I if I wasn't going to, of course, you know, recommend that one, I think that if you are trying to learn more about the foundational aspects of AI engineering in this moment, there's a great book called AI Engineering by Chip Huyen that is also worth taking a look at.

这本书评价非常高，内容非常全面，我也推荐大家去看看。

Very, very well reviewed, a very, very thorough book, and I I also recommend folks check that

好的。

out.

这本书真的非常好。

It's such a good book.

我觉得如果你掌握了其中的概念，那你就已经很不错了。

I I I feel if if you know the concepts there, you're pretty well off.

如果你还不了解这些概念，那你可能还在摸索中。

And if you don't know them, you're probably gonna be, you know, still finding your way.

所以我强烈推荐这本书。

So big plus one for that.

阿迪，这次交流太棒了。

Adi, this has been great.

非常感谢你参与这次对话，这真是一次非常精彩的交流。

Thanks so much for for doing this and this really awesome conversation.

谢谢。

Well, thank you.

我真的很感激。

I I really appreciate it.

我希望，如果你对这个领域感兴趣，你知道的，《超越Vibe编码》现在已经由O'Reilly出版了。

I hope the folks you know, if if you are interested in this space, Beyond Vibe Coding is out now from O'Reilly.

希望它能对一些人有所帮助。

Hopefully, it'll be useful to some people.

但和你进行这次对话真的非常愉快。

But it's been a pleasure having this conversation with you.

是的。

Yeah.

而且我一直在读这本书。

And I I've I've been reading the book.

我非常喜欢它对实用性的深入探讨，所以我也非常推荐这本书。

I I really like how it goes into the practicality, so I I can also very much recommend it.

谢谢。

Thank you.

希望你们在与Adios Mani的对话中，享受了超越氛围编码的乐趣（双关语）。

I hope you enjoyed going beyond vibe coding, pun intended, with Adios Mani.

我从这次对话中学到的一点是，作为工程师，我们持续理解大语言模型到底做什么以及为什么如此重要。

One of the things I took away from this conversation is how it's really important that as an engineer, we keep understanding exactly what LLM does and why.

如果我们不理解，就停下来，弄清楚它，然后再继续。

And if we don't understand it, we just stop, understand it, and then move on.

只要你理解了大语言模型，你就是掌控者。

As long as you understand the LLM, you are in charge.

但一旦你停止理解，它就掌握了主导权，而你可能会变得无助，不知所措。

But once you stop understanding, it's in charge and you can kind of become helpless and out of your depth.

如需了解更多关于如何快速掌握AI工程的技巧，请查看我在《务实工程师》上撰写的深度文章，链接已在节目说明中提供。

For more tips on how to get up to speed with AI engineering, check out deep dives in the pragmatic engineer that I wrote, which are linked to the show notes below.

如果你喜欢这个播客，请在你最喜欢的播客平台和YouTube上订阅。

If you've enjoyed this podcast, please do subscribe on your favorite podcast platform and on YouTube.

如果你也为本节目留下评分，特别感谢你。

A special thank you if you also leave a rating for the show.

谢谢，我们下期再见。

Thanks, and see you in the next one.