靠写氛围代码赚钱：揭秘AI时代的新职业 | Lazar Jovanovic（专业氛围编码师）

尽管英语是第一大编程语言，Lovable 和其他所有工具仍然最擅长通过代码进行沟通。

Even though English is the number one programming language, Lovable and all other tools still communicate in code the best.

如果你想获得像素级精确的结果，就直接给它们代码。

If you wanna get pixel perfect results, just give them code.

它对代码的理解会比你用这些工具时使用的英语、西班牙语或其他任何语言都更好。

It it will interpret it better than your English or Spanish or whatever language that you use in these tools.

所以这是第三种方法。

So that's the third way.

你会想，好吧，现在我更加谨慎了。

You're like, okay, now I'm even more deliberate.

我甚至不再泛泛地给出模糊的概念。

I'm not even going as wide as, like, giving it vague concepts.

我直接提供代码片段。

I'm giving it code snippets.

我希望得到这个精确的设计。

Like, I want this exact design.

我想要这种确切的功能。

I want this exact type of functionality.

所以这就是你的第三个项目。

So that's your third project.

当你完成这三个项目后，你已经达到了一种清晰度，这种清晰度是你如果只是对着一张白纸发呆，或者只是和ChatGPT闲聊而不采取行动时所无法获得的。

And then by the time you do all of these three, you're already at a level of clarity that you wouldn't have if you just sat with an empty piece of paper or maybe maybe chatting just with Chad GPT, but not taking action.

我觉得现在采取行动是如此简单，而且还是免费的。

I think taking action is so so cheap these days and free by the way.

我提到的所有工具都有免费计划。

Like all the tools I mentioned have free plans.

大多数情况下，你根本不用花一分钱，只需启动多个项目就能做到，你猜怎么着？

Like most times you would be able to do this without spending any money at all Just by starting multiple projects because guess what?

这也不会产生额外成本，除了可能消耗一些构建积分外，其他什么都不用花。

That doesn't also cost anything either or doesn't incur additional cost except for builder credits.

你会得到三、四、五、六个不同的方案供你比较。

You're gonna get three, four, five, six different concepts that you can compare.

当你在比较它们时，清晰度会不断涌现。

As you're comparing them, clarity just keeps coming.

事情变得越来越容易理解。

And things get better and better to understand.

而且你也在解决你提到的一个大问题。

And you're also solving for one big problem that you mentioned.

你用了‘AI垃圾’这个词，我很喜欢，因为很多人说‘AI垃圾’时，并不是指美化代码，而是指美化设计。

You used the term AI slop, and I like it because a lot of people, when they say AI slop, they don't refer the a cool beautifying the code, but beautifying the design.

对吧？

Right?

我刚刚提到的这个过程能为你提供四到五个不同的设计选项。

This process that I just mentioned actually gives you four or five different design options.

从长远来看，能为你节省大量积分。

And in the long run save you massive amounts of credits.

因为很多人沉迷于这样的想法：当我给他们这个技巧时，他们会说：‘但这不是更贵吗？’

Because a lot of people obsess over the concept of oh when I give them this this hack they're like oh but that doesn't that cost more?

我觉得是的，从一开始可能确实会多花一点成本。

I'm like yes, upfront, it may cost a little bit more.

但从长远来看，如果你真的想完成这个项目，你实际上能节省数百个积分，甚至可能节省数百美元，更别提节省下来的大量时间了，这一切都源于你从更高的清晰度和更完善的流程开始。

In the long run, if you really want to finish this project, you're actually saving hundreds of credits and maybe even hundreds of dollars, not to mention the amount of days, simply because you started from a point of better clarity and better refinement process.

所以，这就是解决清晰度问题的第一步。

So like, that's the first step of solving for clarity.

还有更多步骤，对吧？

There are more, right?

这是第二层，但我猜你可能对这一点有些疑问。

Which is the second layer, but I assume you may have some questions on this one.

不仅有疑问，而且太棒了，这充分展现了没有工程背景的人进入这个领域所能带来的力量。

Questions and also just wow, this is so such a great, it shows you the power of having someone come into this world without an engineering background.

这种并行构建五次的建议，就是让AI尝试各种可能性。

This advice of just build it five times in parallel, you ask AI to try all kinds of stuff.

这根本不是软件工程师、产品经理或设计师会采用的方式。

Like this is not how someone that has been a software engineer or PM or designer would approach stuff.

所以你的建议非常有趣，就是在开始一个项目时，直接同时尝试五种不同的方法。

So your advice here, which is so fun is, as you're getting started with a project, just run five different approaches at it to start.

第一种是头脑风暴。

One is just brain dump.

这是我正在想的。

Here's what I'm thinking.

这是一个大致的想法，比如你使用Whisper Flow或者内置麦克风。

Here's a general idea, like you use whisper flow or use the built in mic.

然后第二种是，好吧，我现在有了一个大致的想法。

And then two is okay, now I have a general idea.

让我试着把它打出来，真正地思考一下提示内容。

Let me try to type it out, like actually thinking through the prompt.

第三种是，去网上找一个模拟设计，你推荐的网站是Mobin和Dribbble，这两个是你常去的吗？

Three is let me find a mock design somewhere online, and the sites you suggested were Mobin and Dribbble, those are the two that you go to?

是的，大多数时候，好的。

Yeah, most times, Okay.

第四步，这些方法都是并行进行的，非常好：找一个和你想构建的东西相似的实际代码模板，下载那个压缩文件并附加它。

And then the fourth, and these are all in parallel, it's great, Is find like actual code template that looks similar to the thing you want to build, download like the zip file basically and put attach it.

还是只是HTML和CSS？

Or is it just HTML and CSS?

你拿到的就是这种类型的东西吗？

Is that kind of what anything anything you got?

是的。

Yeah.

给你。

Here you go.

好的。

Okay.

然后，太好了。

And then cool.

这是这里的提示。

Here's the prompt here.

帮我做出我想要的东西。

Make me what I want.

我喜欢的是，这里有双重好处。

And what I love is there's two wins here.

一是它能帮你澄清想法，当你看到工具构建出结果时，你会意识到：不对，这不是我的意思。

One is just it helps you clarify the idea as you see the tool build it like, no, that's not what I mean.

让我再试一次。

Let me try it again.

二是你提到的，你可以选择正确的方向，这样就不会被最初的设计和架构束缚住。

And then two is you pointed out, you can pick the right direction, so that you're not locked into your first design and first architecture.

正如你所说，如果你花大量时间去微调设计和方向，那这些令牌就都浪费了。

To your point, if you then spend all this time trying to fine tune design and direction, it's like all these tokens are being lost.

你本可以直接重来。

You could have just started over.

这真是太棒了。

This this is so great.

有人可能会想，当然了，你只是让我们花掉这些可爱的令牌。

Someone may think, okay, of course, you're just getting us to spend all these lovable tokens.

这正是一个可爱的人会告诉我的话。

This is what a lovable person would tell me.

但我真正感受到的是，你在这里能节省最多的钱，因为如果一开始就能做对，就能省去大量试图把它改回你想要状态的工作。

But I what I'm feeling is this is where you could save the most money because if you get it correct in the beginning, you save so much work trying to get it back to where you want it to go.

我确实在帮助人们，这一点毫无疑问。

A million percent that I'm actually saving people.

实际上，我正在违背我本该说的话。

Like, I'm I'm actually going against what I should be saying.

如果我只考虑‘可爱’，我会说：不，不，永远试着去修复它，但那不是我们的业务所在。

If I was thinking about lovable, I'll be like, no, no, just try to fix it in perpetuity but that's not we're not in business of doing that.

我们的业务是赋能任何人构建他们想要的任何东西，而这也是我个人的使命，它深深打动了我——如果没有‘可爱’，我可能这辈子都不会去构建任何东西，我不认为那样的人生会有趣。

We're in business of empowering anybody to build anything that they they want and then, you you know, it's my personal mission that resonates with me because if if there wasn't lovable, I would have never built anything potentially in my life and I don't I don't think that that would have been a fun life to live.

所以，我向人们保证，我已经用这个框架测试过很多人，每个人都告诉我同样的事情。

So, you know, I I guarantee people, like, I've tested this framework with many people and everybody It's telling me the same thing.

令人豁然开朗。

Eye opener.

正如你所说，简单却反直觉。

So simple yet unintuitive as you said.

虽然对我来说，有点说不清道不明。

Even though for me, it's kind of, I don't know.

正如你所说，我把这归因于非技术背景。

As you said, I I attribute it to non technical background.

对我来说，那是我首先会做的事。

To me, that was the first thing that I would do.

我就直接做了。

Like, I just did it.

我从来没想过，哦，我在开发一个了不起的技巧。

I never thought about it like, oh, I'm developing this amazing hack.

我只是觉得，我得等这些代理完成这么长时间。

It I was just like, I'm waiting all this time for these agents to finish.

我还不如开始另一个项目，再一个，再一个。

I might as well start another project and another one and another one.

这其实也是一个提高效率的技巧。

And it's also a productivity hack.

当人们问我，天啊，你怎么能推出这么多东西时？

Like, that's when people ask me, wow, how do you ship so many things?

我会说，我从来不是一次只做一个项目。

I'm like, I never built just one project at a time.

我一次做五六个。

I built five or six.

我有六个可爱的标签页，只是在它们之间来回切换。

I have six lovable tabs, and I just switch between them.

接下来我想聊的另一个技巧是，如果你允许的话，就是反过来的问题：你是怎么进行上下文切换的？

And that's the next hack that I wanna talk about, if you allow me, which is the the question in return is the obvious one which is how do you context switching?

你总在谈上下文，却一直在不同应用之间来回切换。

Like you talk about context so much yet you're a keep switching between apps.

你如何管理这一切，并且以一种高效的方式完成？

How do you manage to do it and do it in a way that's productive?

你怎么做才能避免写出糟糕的代码或推出劣质的产品？

Do you're and not produce bad code or bad product.

这就是我解决LLM问题的方法。

And that's how I solve for that LLM problem.

再说一遍，就像阿拉丁和魔法夹子那样，如果token窗口有限，我该如何让它动态调整？

Again, the Aladdin and the Magic Clamp and all that, which is if there's a limited token window, how do I make it dynamic?

我所说的意思是这样。

And what do I mean by that is this.

如果你只是不断地提示、提示、再提示，你会发现，无论使用什么工具，内存都不是无限的。

If you just go and you prompt and you prompt and you prompt and you prompt, you'll realize that no matter what tool you use, the memory just isn't infinite.

对吧？

Right?

当你到达第10条、15条、20条、30条、40条消息时，早期消息的某些内容就会在对话中逐渐丢失，因为智能体在追求速度，对吧？

By the time you reach message number ten, fifteen, twenty, thirty, forty, snippets of early messages sort of get lost in the translation because agent is optimizing for speed, right?

如果它必须阅读整个对话和你发出的所有请求流，那么开发任何可行或大型项目都将变得不可能，因为这会消耗大量时间、内存和令牌。

If it had to read the entire conversation and the entire stream of requests that you made, developing anything viable or large would be impossible because it's just like consuming a lot of time and a lot of memory and a lot of tokens.

所以，我在早期构建时就意识到，如果它无法记住事情，我的任务就是为它提供参考信息。

So, again, something that I just figured out very early on as I was building was like, okay, if it can't remember things, my job is to provide it with reference.

因此，我会把Lovable或其他任何工具视为一名工程师，我需要在项目进行过程中持续提供上下文。

So let me treat Lovable or any other tool as an engineer that I'm supposed to be providing perpetual context as the project goes.

你可以用很多方式做到这一点，但我发现最高效的方法是进行四个并行构建，我们继续以这个例子为例。

And you can do that in many ways, but the most efficient way that I found was like I would do the four parallel builds, like let's continue off of that example.

当你像我一样构建了数百个项目后，很快就能看出哪个方案是最佳的。

Very quickly after you've built hundreds of projects like I did, you see the winner.

最佳方案如此明显，根本算不上竞争。

The winner is so obvious it's not even a competition.

你可能只需要一两个提示来校准它，一旦确定了最佳方案，我就会向我正在使用的工具提问，或者比如转向ChatGPT或其他工具，让LLM生成一系列PRD。

You maybe do one or more two prompts to calibrate it and when you're like okay the winner is here, at that point, I either ask the tool that I'm using, or I'll maybe, let's say, go to ChatGPT or whatever and ask the LLM to produce a series of PRDs.

PRD是什么？对于不熟悉这个术语的人，PRD是项目需求文档，而对我来说，我称之为‘真相来源’。

What PRDs are for, again, people that are not familiar with the terms, are project requirements documents, or for me, I call them, like, sources of truth.

对吧？

Right?

从几个角度来看，观察这个项目成功所必需的前提条件。

Watch what needs to be true for this project to be successful from a couple of perspectives.

我通常会构建一个我称之为总体规划的东西。

I usually build something that I call a master plan.

这就像一个指南针，指出我们正在构建什么，对吧？

It's basically a compass saying here's what we're building, right?

这就像在和一个人交谈。

It's like talking to a human.

我真的很把Lovable当作一个人来对待。

I really treat Lovable like a human being.

所以这就是我们正在构建的东西。

So it's like this is what we're building.

然后我会制定一个实施计划，说明我们将如何构建它，以及具体的步骤顺序。

Then I build an implementation plan which is this is how we are gonna build it in this is the sequence.

对吧？

Right?

再次强调，质量、品味和人性对我来说非常重要。

It's very important to me again, going back to quality, taste, human nature.

我需要明确界定，因为我仍在使用一个尚未具备情感智能的系统。

I need to define because I'm still working with a system that is not emotionally intelligent yet.

我需要定义我希望这个应用的外观和感受是什么样的。

I need to define how I want the app to look and feel.

所以我还会制定另一份产品需求文档，即设计规范。

So another PRD that I build is design guidelines.

最后，还有一个能将所有这些串联起来的方面，那就是：好吧。

And then finally, something that just circles it all around, which is like, okay.

当我们知道事物的外观，也清楚如何构建它时，用户界面应该是什么样子？

When we know how things look and when we know how we're building it, how does the user unit look like?

对吧？

Right?

我用注册表，然后呢？

I use a registers and then what?

当他们注册并完成第一步后，第二步是什么？

And then when they register and do that first step, what's the second step?

第三步是什么？

And what's the third step?

还有其他各种步骤。

And whatnot.

所以我至少写了四个产品需求文档，对吧？

So I built at least four PRDs, right?

当这些文档完成后，我会阅读它们。

And then when these are built, I read them.

这就是规划讨论的部分。

That's the planning chatting part.

我现在会花很多时间在这上面。

Like that's where I'll spend a lot of time now on.

当我敲定第一个设计方案时，如果需要，我会花一整天时间来规划这一部分，比如做文档和拆解任务，因为这就是我设定方向的方式。

When I nail down that first design, I'll spend an entire day if I need to just planning this part out, like documentation and breaking things down because that's how I'm setting the course.

整个流程的后续所有内容都将依赖于这一环节。

Like everything's gonna be dependent on on this particular part of the process.

当我完成这些后，我会创建一个最终文档，我称之为计划。

When I'm done doing that, I build one final document, which I call either plan.

Md 或者任务。

Md or tasks.

Md 中的 .md 部分，你知道，就是 Markdown。

Md and .md part is, you know, markdown.

我之所以使用 Markdown 格式，是因为我发现 AI 更喜欢阅读 Markdown。

Basically, I'm just using markdown format because I've learned that AI likes to read markdown.

这个文档充当了真实任务和子任务的权威来源，AI 需要根据这些任务执行操作才能达成目标。

And what that serves is a it serves as a source of truth on like actual tasks and subtasks that it will need to execute to get to the finish line.

对吧？

Right?

然后还有最后一层，取决于你使用的工具。

And then there's the final final layer which is depending on what tool you use.

Cloud Code 或 Cursor 有被称为规则的东西。

Cloud Code or Cursor have what's known as rules.

Md 或代理。

Md or agent.

Md。

Md.

你使用规则或代理文件时，实际上是在告诉代理你希望它如何行为，以及长期应关注什么，这样你就不用在每次提示时都重复说明。

What you're basically doing with rules or agent files is you're letting the agent know how you want it to behave and what it should focus on in the long run so that you don't have to repeat yourself with every prompt.

对吧？

Right?

所以在 Lovable 中，项目设置里有一个单独的菜单，你可以在那里定义项目知识。

So in Lovable, there's a there's a separate menu for that in your project settings where you can define project knowledge.

通常我会说：嘿，在做任何事情之前，先读取所有文件。

And usually what I'll say, hey, read all the files before you do anything.

比如在阅读所有产品需求文档和任务之前，不要做任何操作。

Like don't do anything before you read all the PRDs, read tasks.

查看MD文件了解下一个任务是什么，然后执行那组任务，完成后告诉我你做了什么以及我应该如何测试。

Md to see which task is next, then execute on that next set of tasks, and when you're done, tell me what you did and how I should test it.

这就是为什么我会认真阅读代理输出的对话内容开始发挥作用。

And that's where that conversation about I religiously read the agent output comes into play.

我已经告诉代理，我提供了它成功所需的所有工具和资源。

I've told the I gave the agent everything, all the tools and resources that it needs to succeed.

我给了它规则。

I gave it the rules.

我给了它文档。

I gave it the docs.

我告诉它应该如何处理这些内容。

I told it what to to do with them.

到了那个阶段，我就只是坐着阅读了。

And at that point, I'm just sitting and reading.

哦，我不再发提示了。

Oh, I don't prompt anymore.

从那以后，我可以随意切换任意多个窗口。

At from that point on, I can switch as many windows as I like.

我的提示已经变成了‘继续下一个任务’。

My prompts have become proceed with the next next task.

我不需要上下文了。

I don't need the context.

我把这个任务外包并委托给代理了。

I outsource that and delegate that to to the agent.

代理需要上下文，而我需要确保它是动态的。

The agent need needs context, and I need to make sure that it's dynamic.

我需要定期更新文档，以便随着时间推移调整它所使用的令牌窗口及其使用方式。

I need to make sure that I'm regularly updating the documents from time to time so that we shift that token window it it uses and how it uses it over time.

但我不会再发提示了。

But I'm not prompting.

我没有打断流程。

I'm not interrupting the flow.

是的。

Yes.

我会进去测试，偶尔在这里或那里加点提示，但正是这样，我才能同时推进五个项目，而且从不丢失生产力——正如我所说，我现在是手动完成这些的。

I'll go in, test, maybe put a prompting here or there, but that's how I can build five projects simultaneously and never lose the productivity part, which is, again, as I said, I do this today manually.

三个月后再打电话给我。

Call me to talk three months from now.

到时候会有一个代理帮我做这些。

An agent will do this for me.

我差不多就要失业了。

I'll be out of job pretty much.

这就是为什么我完全不优化这项技能。

That's why I don't optimize for this skill at all.

我现在用它来绕过人性和大语言模型的缺陷，但我今天100%的时间都花在了良好的判断力、清晰度、质量、品味、优秀的文案和合适的字体上。

Like, I'm using it today to bypass the shortcomings of human nature and LLMs, but I'm optimizing a 100% of my time today on good judgment, clarity, quality, taste, good copy, good fonts.

人们根本不会谈论那些与AI配合使用的字体。

Like, people don't talk about fonts at all that work with AI.

在我看来，字体至少占了60%，甚至可能更多，决定了你的输出看起来怎么样。

They're like 60% in my mind, maybe even more in how your output's going to look like.

这才是我的执着。

That's my obsession.

我不纠结于我今天提到的这些事情，因为我清楚未来会怎样。

Like, I don't obsess over these things that I'm talking today because I know what's coming.

智能代理会变得越来越好。

Like, the agents are gonna get better.

模型也会变得越来越强大。

The models are gonna get better.

它们不再需要我来扩展上下文。

They're not gonna need me to extend the context.

它们会自己完成这些。

They're gonna do it themselves.

所以对我来说，我所优化的技能是那种需要更好决策能力的技能，而不是更好的输出或更好的对齐。

So for me, the skill that I optimize for is is the the one that that, like, requires better decision making rather than better output or better alignment.

天哪。

Oh my god.

这里的内容太多了。

There's so much here.

这太棒了。

This is so awesome.

好吧。

Okay.

所以本质上，这里发生的是：你启动一个项目，尝试各种方法，然后选择一个感觉最正确的方向。

So essentially what what's happening here is you start a project, try a bunch of stuff, pick a direction that feels most correct.

一旦确定了方向，你就会花大约一天的时间，不是去构建，而是与这个AI代理一起进行规划。

And once you have a set direction, you spend essentially a day not building, but working with this AI agent to plan.

然后，这就是我想谈谈的原因。

And then, and why I want to talk about that.

一旦你有了计划，就会发现，用一些人可能认为并不复杂的工具，竟然能构建出如此强大的东西，这真是太神奇了。

And once you have the plan, then it's, and it's amazing that you could do stuff like this with what people may, some people may feel are not sophisticated tools that can build incredibly powerful things.

你可以用像Lovable这样的工具完成很多这类工作，比如制定计划、规则和MD文件。

Like you can do a lot of this with tools like Lovable, like have plans and rules and MD files.

很多人可能并不了解这一点。

You know, a lot of people may not think, may not know that.

所以核心思路是：花大量时间做规划，因为这能为你节省后续的大量时间，只有在有了明确计划之后，你才开始推进。

And so the idea is, okay, spend all this time planning, because again, that'll save you a lot of time down the road, and then only once you have a plan, you have what you get it going.

这其中一个关键部分是‘三个愿望’规则，这一点非常重要。

And a key part of this, this three wishes rule is really important.

你这么做，很大程度上不仅是为了让计划更加清晰，更是因为一次只处理一个任务，能保持代理的上下文窗口较小，避免它迷失方向。

The reason you're doing this in large part beyond just being really clear about the plan is this idea of one task at a time keeps the agent's context window small, so that it doesn't lose track of where it's at.

这部分看起来很重要，对吧？

That part seems important, right?

就是去做这件事。

It's like do this thing.

然后，好的，太棒了。

And then, okay, cool.

现在进行下一步。

Now do the next step.

对吧？

Right?

是的。

Yes.

对。

Yes.

因为Again，如果你没有这么做，我们来谈谈你忽略这一点的情况。

Because again, if let's say you didn't do this, let's talk about you ignoring this.

你会说，我就想随性而为。

You're like, I just want to vibe my way.

好的，很棒。

Okay, great.

没问题。

No problem.

你工作，你工作，你工作。

You work, you work, you work.

总有一天，出问题了，对吧？

At one point, something breaks, right?

你什么都没记录。

You haven't documented anything.

没有任何参考点。

There's no reference points.

你报告了一个问题。

You report a problem.

你根本没参考任何文件或架构。

You're you're not referencing files or architecture at all.

你只是在描述问题。

You're just describing the issue.

接下来会发生什么。

Here's what's gonna happen.

任何工具，无论是Lovable、Cursor还是Clot，你提到的任何工具都会这样。

Any tool, Lovable or Cursor or Clot, whatever tool you talk about is going to do this.

它会说，好吧，让我开始调查一下。

It's going be like, okay, let me start investigating.

然后你的代码库会变得越来越大、越来越大、越来越大。

And then your code base gets bigger and bigger and bigger and bigger and bigger.

当你刚开始时，你只有大约20个文件。

Like when you first start, you have like 20 files.

它可以读取20个文件。

It can read 20 files.

但当你现在正在开发一个项目，里面有六七十个边缘函数时，会发生什么？

But what what happens when you have I'm just building a project right now that has like sixty, seventy edge function functions.

对吧？

Right?

那当我发现出问题了，却不知道哪个边缘函数是做什么的，会发生什么？

What happens then when I say this broke and there's no reference which edge function does what?

猜猜看？

Guess what?

Lovable 会读取所有这些文件。

Lovable is gonna read all of those.

它会消耗80%的令牌配额用于阅读以获得清晰理解，只留下最后的20%用于思考和执行。

And it's gonna consume 80% of the token allocation on reading to get clarity, leaving only the final 20% for thinking and executing.

我的猜测是——我无法证明这一点，评论区的LLM专家可能会说我是错的，但作为非专业人士，这是我最好的推测。

What I'm guessing, and I can't prove this, an LLM expert in the comments may say that I'm wrong, but this is my best guess as a non educated person.

这些家伙非常顺从，也非常容易被说服。

These fools are very obedient and very agreeable.

它们会对你撒谎。

They're gonna lie to you.

它们会告诉你问题已经修复了，即使根本没有。

They're gonna tell you that they fixed the problem, even though they didn't.

它们只是会试图让你感到开心，说：是的，我找到问题了，我已经解决了。

They're just gonna try to make you feel happy and say, yes, I found what the problem is and I fixed it.

但很多时候，它们其实并没有解决，人们却怪罪机器。

When a lot of times when they don't, people blame the machine.

在某种程度上，我承认这确实没错。

And to to an extent, I I will say that's true.

这是你的错，朋友。

It's your fault, my friend.

你没有给这个工具提供任何清晰的信息或上下文。

You did not provide any clarity or context to this tool.

你只是盲目使用它的原始能力，结果像车轮打滑一样，越陷越深。

You just used its raw power and dug a deeper hole with your spinning your wheels into the into the mud.

对吧？

Right?

而且，你知道，显然，我认为我们正走向一个AI更诚实而非一味顺从的世界，比如会说：嘿。

And and, you know, obviously, I think we're heading into a world where AI is more honest than obedient and saying, hey.

我只是部分修复了这个问题。

I only partially fixed this.

你知道的。

You know?

你没有给我足够的上下文。

You did not give me enough of a context.

对吧？

Right?

人们接下来更大的错误就是，相信这个工具已经修复了它。

The bigger mistake that people make then is like, they trust the tool fixed it.

他们测试了一下。

They test.

他们发现没修好，就开始生气、骂人，就像我们说的那样，然后情况变得更糟，因为猜猜怎么着？

They see it didn't then they get mad at it, start cursing and yelling, as we say, and then it gets even worse because guess what?

AI的另一个缺点是：尽量不要伤害你的感受，绝不会说你是个傻瓜。

Another bad trait of AI is: best not to hurt your feelings and never say you're the dumb one.

它说：不，我是笨蛋。

It says no, I'm the dumb one.

所以在下一次请求中，它没有专注于阅读，而是又花了30%的token来想如何道歉。

So it focuses in the next request instead of focusing on reading, it spends another 30% of tokens trying to come up with an apology.

对吧？

Right?

我虽然没受过什么教育，但如果你读过像ChatGPT这样的思考模型的输出流，你就完全明白我的意思了。

Again, I'm not educated but that's if you ever read like a stream of ChatGPT's thinking in thinking models you see exactly what I mean.

比如当我责备它时，我看到第一条消息就是：好吧，用户生气了，我得想想怎么缓解它的焦虑之类的。

Like when I insult it I see that the first message is, okay, the user is mad, so I need to think of ways how to reduce their anxiety or whatever.

我心想：天啊，我真是中了最糟糕的招了。

I'm like, oh man, just fell for the worst drink in the book.

我让它把最稀缺的资源——那些token——花在思考如何安抚我的情绪上，而不是专注于解决实际问题。

I made it spend the most scarce resource, which is those tokens, on thinking how it should address my anxiety versus focusing on the actual problem.

所以我的建议是：没错，你可以随意互动来取乐，也可以在原型设计时随意互动，因为那属于探索阶段。

So my advice for people is like, yes, vibe your way for fun and vibe your way while you're prototyping because that's exploration part.

我喜欢这一部分。

I love that part.

但当探索完成后，请务必使用参考文档。

But when exploration is done, please, please, please use referencing documentation.

请尽可能使用所有代理文件，因为这些令牌分配非常稀缺。

Use all the agent files that you can because DAT that token allocation is so scarce.

随着时间推移，这些功能会逐渐扩展。

Like, it's gonna get expanded over time.

事情会变得越来越便宜、越来越快，但目前它们依然非常宝贵。

Things are gonna get cheaper, faster, but right now it's still so valuable and precious.

你必须确保它们被用在正确的地方。

You really need to make sure that they are allocated in the right direction.

这太好笑了。

This is hilarious.

我觉得这个精灵的比喻特别贴切。

I think the the genie metaphor is so good here.

只是想想这个精灵，你就得清楚自己到底想要什么。

Just thinking about this genie is you're trying to be clear about what it is you want.

如果你只是随随便便地‘感觉’许愿，它就会做错事。

And if you're just like vibe, you know, vibe wishing, it'll do the wrong thing.

所以这里的建议是，尽可能多地提供关于你希望它做什么的上下文。

So the advice here is be as give it as much context about what you want it to do as possible.

这些文件我们稍后会详细讨论。

And these files we'll talk about right after this.

但这里的理念就像激光一样，精准地聚焦在你希望它解决的问题上。

But the idea here is just like laser show the point, laser at where you want it to fix the problem.

别以为它自己会去搞清楚，因为它真的会去尝试，而且会浪费你所有的令牌，填满上下文窗口。我记得在这次录音前你提到过，当上下文窗口空间快用完时，解决方案就突然出现了。

Don't just assume it'll go figure out because it will, and it'll try really hard to, and it'll waste all your tokens, it'll fill the context window, and I remember at one point you mentioned before this recording that because it starts to run out of space in the in the context window, it start it just like the solution ends up.

它最终并没有真正花大力气去解决问题，因为它把所有精力都花在了阅读和思考上，然后到了最后一刻，它就说：好吧，这就是解决方案。

It doesn't actually work that hard on figuring it out in the end because it spent all this energy on reading and thinking, and then it's like, okay, here at the last second, here's the solution.

我觉得它只是挑了第一个它认为出问题的地方。

I think it just picks the first thing it thinks is broken.

这再次说明，我完全是门外汉，刚加入对话，只是在随口思考。

That just again, this is me completely uneducated coming in into the conversation and just thinking out loud.

这只是我的直觉，也是我逻辑上思考问题的方式，就是说，嘿。

That's just my gut feeling in the way I think logically about it, which is, hey.

如果它消耗了大部分上下文窗口，并且知道它快用完了，也许它意识到自己快用完了，也许没有，但无论如何，我有过一些亲身经历：我的请求不够清晰。

If it consumes most of its window and knows that it's running out of it, maybe it's aware that it's running out, maybe it isn't, but either way, I've had the experience anecdotally to where, like, my request is unclear.

我觉得它只是选择了最简单的解决办法，而不是相反的情况——我花大量时间寻找正确的文件，引用该文件，费力地引导它，甚至给它一盏手电筒，然后说：问题就在这里。

I feel it takes the easiest fix in the book, just the easiest versus the other way around where I'm, like, spending so much time finding the right file, referencing that file, like, putting in the effort of hand holding it in dark, maybe giving it a flashlight, and then saying, here's the problem.

我认为这个文件才是问题所在。

I think that this is the problematic file.

然后它就会说：哦，对。

And then it's like, oh, yeah.

你说得对。

You're right.

现在我要一遍又一遍地真正去修复。

And now I'm gonna actually fix over and over and over.

我见过这种情况，因为说到底，我只看输出结果。

And I've seen that because, again, all I do is read the the output.

代理让我学会了如何使用它。

Agent makes me learn how to use it.

所以人们看的东西我不知道，但我只看输出。

So people read I don't know what people read, but all I read is the output.

比如，我不看代码，那是以后的事了。

Like, I don't read the code and it's later down the road.

因为我知道，它在这方面比我强得多。

Because, like, I know that it can do that much better than I can.

再说一遍，我记得读过一句很棒的话，但抱歉，我一时想不起作者是谁，不过大意是：AI的上限不在于模型的智能。

Again, I feel if there's a good quote I've read, I can't I apologize to the author because I can't attribute it off the top of my head, but it's like, the ceiling on the AI isn't the model intelligence.

而在于模型行动之前所看到的内容。

It's what the model sees before it acts.

对吧？

Right?

所以这就是目前的上限。

So that's the ceiling right now.

你的代理暴露在哪些内容里呢？

Like, what do what are you exposing?

我们常谈到人类的暴露时间。

We talk about exposure time for humans.

在代理进行代码修改之前，你让它接触什么内容同样重要，甚至可能更重要。

What are you exposing your agents to as well is as important, if not even more important, before it makes code edits.

是的。

Yeah.

回到这些文件，我觉得这真的很重要。

Coming back to these files, I think this is really important.

那么，对于想要更好完成这件事的人，最简可行方案是什么？

So let's think about just like what's like the MVP for someone that wants to do this better?

你列出了所有这些文件，本质上都是MD文件，你在一天中开始实际构建之前就一直在构建它们。

You listed all these kind of file these MD files essentially that you're building over the course of a day before you start actually building the thing.

你有设计指南、用户旅程、任务、代理MD文件、规则MD文件。

You had design guidelines, the user journey, tasks, agents MD, rules MD.

假设你只想进一步前进，把这件事做得更好。

Say you wanted to just, like, move one step forward and be better at the stuff.

你会创建哪些文件？

What are the what are the files you'd create?

然后这些文件大致是什么样子？

And then what do they roughly look like?

这些文件里面包含什么？

What's inside these files?

是的。

Yeah.

首先是总体规划，它是一个万米高度的概览，对吧？

So the master plan is the first one, which is like, it's it's a 10,000 foot overview, right?

它非常宏观地解释了我对这个应用的意图。

It it it really high level explains the intent that I have with this app.

你知道的吧？

You know?

主计划就是那个MD吗？

Is master plan that MD.

你是这么叫它的吗？

Is that what you call it?

是的。

Yes.

对。

Yeah.

主计划那个MD。

Master plan that MD.

好的。

Okay.

它其实就是说，嘿。

And it's like, it's really just hey.

这就是我这么做的原因。

This is why I'm doing this.

这就是我为谁做的。

This is who I'm doing it for.

这就是我希望他们让你感受到的。

This is how I want them to to you feel.

在主计划中，我经常引用其他的PRD。

And a lot of times in the master plan, I will reference the other PRDs.

我会说，设计需要显得现代且精致，但关于具体的参数，请参考并阅读设计规范。

I'll be like, the design needs to feel modern and slick, but for exact, you know, parameters, consult and read design guidelines.

主计划。

Md.

对吧？

Right?

所以我只是把主计划当作一个高层次的概览，对吧？

So I'm using just the master plan as like this high level overview, right?

为了让代理了解，哦，好的，我们正在构建 x、y、z。

To get the agent into, oh, okay, yeah, we are building x, y, z.

对吧？

Right?

然后是实施计划，因为毕竟需要有一定的顺序。

Then there's the implementation plan because, you know, there needs to be some order.

如果你只是随意堆叠东西而不讲顺序，永远都到不了终点。

If you just like dump stuff on top of each other without any order, you're never gonna get to the finish line.

这就是任务。

And this is Tasks.

对吗？

Md?

你管这个叫这个吗？

Is that what you call this?

不是。

No.

那就是实施计划。

That that's the implementation plan.

实施，是的。

Implementation Yeah.

好的。

Okay.

实施计划在某种程度上是为了服务未来的任务。

And implementation plan is kind of in service of the future tasks.

嗯。

Md.

如果所有这些文件都是为了构建任务服务的。

If that all of these files are in service of building tasks.

嗯。

Md.

当你构建任务时。

When you build tasks.

嗯，其余的部分几乎无关紧要。

Md, then the rest is almost irrelevant.

它只是你用来构建和执行任务的基础，对吧？

It's just the basis for you to build tasks to execute, right?

实施计划是第一层，同样也是一个更高层次的概览。

The implementation plan is kind of the first layer, which is again higher level overview.

它不会深入探讨如何实现，只是解释说：如果我们正在构建这个，我认为我们应该先从后端开始。

It doesn't go into the depth of like how to get there, it just goes into the explaining of like, oh, well, if we're building this, I think we should start with the back end.

我们应该先从数据库表开始，然后再处理身份验证。

And we should start with tables and then later authentication.

之后，我们会引入API。

And then after that, we're gonna bring in the API.

再之后，我们就做这个。

And then after that, we're gonna do this.

这again，只是想一下，我是个擅长出主意的人。

It's again, just think of it as having I'm an ideas guy.

我正和一个技术专家坐在一起。

I'm sitting with a technical guy.

就是你和我。

It's me and you.

我们正在打造我们的初创公司。

We're building our startup.

你知道，你有软件工程背景，而我正在向你讲述我的想法，对吧？

You, I know you're a software engineer by background and I'm telling you my idea, right?

我给你整个总体规划，然后你回来告诉我：好的，如果你想这么做，是可行的。

I'm giving you the master plan and you come to me back and you're like, okay, if you want to do this, it's doable.

这是我建议的顺序。

Here's how I would order it.

你根本没有路线图。

Like, you don't have a road map.

你根本没有打开你的线性流程，就开始写功能和RFC之类的了。

You you're not you didn't open your linear and started writing features and and RFCs and whatever.

你只是从宏观层面谈论事情的顺序。

You're just high level talking about the the order of things.

然后，作为两位联合创始人，我和你再次讨论，说：好吧，如果我们对此达成一致，那这个东西应该是什么样子？

And then me and you, again, as two co founders, we talk and say, okay, well, if we agree on this, like, how how should this look like?

它应该是什么感觉？

How should this feel?

对吧？

Right?

让我们先从宏观层面描述它，但现在因为我使用了AI，我可以深入一点。

Let's describe it high level, but now because I use AI, I can go a little bit deeper.

这就是我喜欢看到Lovable或其他工具的地方。

And that's where, like, I I like to see Lovable or any other tool.

ChatGPT在这方面很擅长。

ChatGPT is good at it.

我甚至自己构建了一个自定义的GPT，所以如果人们想在使用任何工具之前先有个起点，他们可以去ChatGPT商店搜索GPT，输入‘lovable base prompt generator’或‘lovable PRD generator’，找到我创建的那些，直接把想法丢进去，就能得到这些文件作为输出。

I even have my I built, like, custom GPT, so if people wanna start somewhere before they even get into any tool, they can go to ChatGPT store and for GPTs and just type lovable base prompt generator or lovable PRD generator and find those that I built and just, like, brain dump in them and then get these files as output.

对吧？

Right?

所以我喜欢在设计规范中看到一些CSS元素，因为设计这事儿有点微妙。

So I'll I I like to see some elements of CSS in in design guidelines because, you know, you with design, it's a little bit it's a little bit tricky.

AI有时过于富有创意。

AI is sometimes over creative.

所以这时候我会进行一些更技术性的引导。

So I that's where I'm doing a little bit more technical steering.

对吧？

Right?

最后，就是用户旅程了。

And then finally, it's just the user journeys.

只要我们知道东西长什么样、感觉如何，以及我们高层面要构建什么——再次强调，是非常高层面的。

Just, like, if we know how things look like, if we know how they feel, if we know what we're building high level, like high level, just very high level again.

用户是怎么操作的？

How do people navigate?

这里面有哪些功能呢，你知道的，类似这些东西？

What what are some of the the features in there, you know, and stuff like that?

然后是MD深入细节的任务，比如如果你想要这些用户旅程，并且想先搭建后端，这里有一系列我需要完成的任务。

And then tasks that MD gets into the nitty gritty of like, oh, if you want these user journeys and you want the back end built first, here's a set of tasks that I need to do.

它只是把这当作输入。

Like, it just takes that as an input.

我只是让工具去做那些过去人类要花大量时间做的繁琐工作。

I'm just making the tool do the do the, you know, that gritty work that humans used to spend so much time on.

我觉得有了这些工具，我们每个人都在变成超级产品经理。

Like, I feel like with with these tools, we're all becoming product managers on steroids.

你知道的，我们只是在利用AI，但我觉得好的产品经理，他们的报酬并不是来自写好的产品需求文档。

You know, like, we're just leveraging AI, but like good product managers, I think, are not compensated for writing good PRDs.

他们的报酬，再次强调，来自于良好的判断力。

They're compensated, again, for good judgment.

别人可以负责撰写。

Somebody else can do the writing.

作为负责指导和构建这个产品的人，你必须清楚，到底什么才是有用的？

You, as somebody who directs and builds this product, you need to know, again, what what what's gonna be useful?

什么才是有品位的？

What's gonna be tasteful?

什么才能真正产生影响？

What's gonna be something that actually moves the needle?

不过，我得说一点。

I will say one thing, though.

尽管我反复强调，你需要培养品位。

Just because I put so much emphasis on, like, oh, you need to acquire taste.

但这并不意味着你不该去动手构建。

Oh, you that doesn't mean you shouldn't build.

你只有通过实际去构建，才能变得更好。

You get better at this by building actually.

所以，所有听到这段话的人，现在就去动手做一个东西吧。

So everybody listening to this should like literally go and build something today.

一、二、三、四、五个项目，去测试所有这些工具，因为只有通过实践，而不仅仅是阅读，你才能获得清晰的理解。

One, two, three, four, five projects, test all of these tools because that's how you get to clarity, not just by reading, but also by by doing as well.

我给你出个谜题。

Here's a puzzle for you.

OpenAI、Cursor、Perplexity、Vercel、Platt 以及数百家成功的公司有什么共同点？

What do OpenAI, Cursor, Perplexity, Vercel, Platt, and hundreds of other winning companies have in common?

答案是，它们都由今天的赞助商 WorkOS 驱动。

The answer is they're all powered by today's sponsor, WorkOS.

如果你在为企业开发软件，你肯定体验过集成单点登录、SCIM、RBAC、审计日志以及其他大型客户所需功能的痛苦。

If you're building software for enterprises, you've probably felt the pain of integrating single sign on, SCIM, RBAC, audit logs, and other features required by big customers.

WorkOS 将这些阻碍交易的难题转化为即插即用的 API，专为 B2B SaaS 构建的现代化开发者平台。

WorkOS turns those deal blockers into drop in APIs with a modern developer platform built specifically for b to b SaaS.

无论你是正在争取首个企业客户的种子期初创公司，还是正在全球扩张的独角兽企业，WorkOS 都是让你快速达到企业级标准并实现增长的最快路径。

Whether you're a seed stage startup trying to land your first enterprise customer or a unicorn expanding globally, WorkOS is the fastest path to becoming enterprise ready and unlocking growth.

它们本质上就是企业级功能的 Stripe。

They're essentially Stripe for enterprise features.

访问 workos.com 开始使用，或者直接联系他们的 Slack 支持，那里有真正的工程师会迅速回答你的问题。

Visit workos.com to get started or just hit up their Slack support where they have real engineers in there who answer your questions super fast.

WorkOS 让你能够像顶尖团队一样开发，拥有出色的 API、详尽的文档和流畅的开发者体验。

WorkOS allows you to build like the best with delightful APIs, comprehensive docs, and a smooth developer experience.

立即访问 workos.com，让你的应用今天就具备企业级能力。

Go to workos.com to make your app enterprise ready today.

我想象听到这些的人可能会觉得，这工作也太多了吧。

I'm imagining people hearing this may start to feel like this is so much work.

我得坐在这儿制定所有这些规则，还要弄清楚这么多细节。

I just have to sit here and create all these rules and figure out all these little details.

一方面如此，另一方面又不然。

Like in one sense, is in another sense.

你花几个小时，最多一天做规划，然后让 AI 去构建这个原本需要别人几周甚至几个月才能完成的东西，对吧？

This is like you spend a few hours, maybe a day planning, and then you have AI build this thing that would have taken somebody weeks, months, right?

他们说，实现这件事所需的投资，回报率高得离谱。

They're like, the amount of investment to achieve this thing is absurd ROI.

这也展示了专业级编码究竟是什么样子。

It also this shows you just what professional vibe coding looks like.

你知道，每个人都会想象编码就是坐在那儿打字，去做这个做那个。

You know, everyone imagines vibe coding, I'm just sitting here typing stuff, go and do this.

很好。

Good.

如果你真的想打造一个像你所说的、能产生重大影响、解决人们真实问题、持久且可扩展的优秀产品，这就是你要做的方式。

If you want to actually build something really great that moves the needle as you said, that solves people's real problems, that lasts, that you know, scales, this is this is how you do it.

如果你真的想把这当作一份正经工作，同时也想打造真正出色的产品。

If you really want to do this as a as a real as a job, and also if you wanna build things that are really great.

是的。

Yeah.

别误会我的意思。

And don't get me wrong.

当然，原型设计有着巨大的价值。

Like, there's there's obviously a ton of value in prototyping.

比如，可能有很多观看这个视频的人会想，好吧。

Like, there are a lot of people maybe watching this that are like, okay.

我想在工作中使用Lovable，但我不能，或者各种原因吧。

I wanna use Lovable at work, but I I can't or whatever, you know, there's there's different reasons.

也许你身处医疗或金融行业，有某些监管规定阻止你将产品上线。

There's maybe you're in healthcare or finance or there's something regulatory that just prevents you from pushing to production.

为了原型而构建，本身就是最好的应用场景之一。

Like building for the sake of prototyping is one of the best use cases.

我们2025年的口号是‘演示，别写文档’——与其写一大堆文档、开会、跟工程师反复沟通，试图让市场或销售团队理解你的想法，不如直接进Lovable，三十分钟内做出原型，然后交出去。

We Our motto for 2025 was demo don't memo, which is like instead of writing all these documents and talking and sitting on meetings with your engineers trying to get your vision as a marketer or a sales guy in the office across, go into Lovable and build the prototype in thirty minutes and just hand it over.

我以前确实有一份正经工作，是在加入Lovable之前。

Like and I have a real, like, job that I held before Lovable.

事情就是这么发生的。

That's exactly what happened.

去年这个时候，我需要构建一个真正企业级的产品。

Like, this time last year, I needed something built enterprise grade, really.

当时，Lovable 和我都还没有能力构建它，但我有一支我合作过的工程师团队。

And Lovable and myself were not there yet to build it at that point, but I have I had a team of engineers that I worked with.

我花了四个小时就构建出了原型，他们实际上在六到七个月后成功将其复制到生产环境中，连接了所有系统和组件。

I built the prototype in four hours, and they actually were able to replicate it six to seven months later into production with with connecting all the pipes and everything.

但如果要我描述的话，我会说，光是把想法说清楚，我就得花上至少一到两周的时间。

But, like, if I had to describe it, I would say it took me it would take me at least a week or two just to get the words out there.

我只是坐下来，四个小时就把它做出来了，这就是 Lovable 一月的样子。

I just sat and built it in four hours, and that's, like, lovable January.

而今天的 Lovable，2026 年一月，功能上已经远远领先了。

This lovable today, January 2026, is, like, ages ages ahead with functionalities.

它变得好太多了。

Like it's so much better.

这根本没得比，对吧？

It's not even a contest, right?

所以我认为，以我们现在的阶段来说，据我所知，标普 500 强公司中至少有一半的企业，都有人在使用 Lovable。

So I think now with our stage where like for instance there's I'd say at least to best of my knowledge at least half of S and P 500 companies have people working in them that are using Lovable to some extent.

对吧？

Right?

我们有很多企业公司正在使用Lovable的企业版，创建非常有意义的项目。

And we have a lot of enterprise companies that are actually on enterprise plans with Lovable that are creating super meaningful projects.

我不会点名具体公司，但全球领先的网约车公司、电信公司，以及在医疗、金融等众多领域的顶尖企业，都在积极地让他们的团队使用Lovable。

Like, I'm not going to name names, but like leading rideshare companies of the world, leading telecommunications companies of the world, leading companies of the world in many, many aspects, healthcare, finance, like are actively with their teams using Lovable.

反馈总是相同的：是的，我们可能还不能直接上线，但我们的市场团队不再需要等待工程师了。

And it's always the same feedback, which is, yes, we may not be able to push to prod, but, like, our marketers are no longer waiting for engineers.

像市场、销售、人力资源等岗位的同事，现在都能自信地自行搭建内部工具，用来管理我们的开支、员工入职流程等等，类似这样的用例非常多，你看到Lovable以及其他类似工具正在被用于推动各种功能上线。

Our, you know, people in go to market or sales or HR or whatever roles are now just confidently building internal stuff for us to manage our expenses or manage employee onboarding or like there's so many use cases like that, where like you're seeing Lovable and and other tools for that matter being used to to push things into production.

你知道的吧？

You know?

为了帮助人们实现你所描述的这种工作流，你觉得在录制结束后，能否分享一些简单的模板，让大家看看这些MD文件长什么样，方便直接参考和复制？

To help people do this workflow that you're describing with all these MD files, do you think you could share after we record this just templates, like simple templates of what these files look like for people just to look at and copy?

我会像之前说的那样，直接打开ChatGPT，把想法一股脑儿地丢进去，输入‘lovable GPT lovable PRD生成器’。

I would literally go to ChatGPT as I said and brain dump into it in my just type lovable GPT lovable PRD generator.

你会看到我的名字在那里。

You'll see my name there.

对吧？

Right?

而且我是作者。

And and and that I'm the author.

进去，把想法一股脑儿倒出来。

Go in, brain dump.

它会问你几个问题来明确需求，然后为你生成四个文件，你可以直接上传这些文件。

It will ask you a couple of questions to get clarity and just produce four files for you, and you can just go ahead and and upload those.

太棒了。

Amazing.

不错。

Cool.

我们会附上这个链接。

We'll link to that.

所以这不仅仅是给你一堆文件。

So so it's not it's not just here's a bunch of files.

我们来跟这个工具聊聊吧。

Let's go talk to this thing.

它会为你生成正确的文件，然后你把这些文件接入LevelBull或其他工具。

It'll generate the right files for you, and then you plug that into LevelBull or other tools.

是的。

Yeah.

它经过训练，能够像我一样思考。

It's trained on it's trained to think like I do.

所以没错。

So yeah.

哦，太棒了。

Oh, amazing.

好的。

Okay.

这太完美了。

That is perfect.

顺便说一下，你想聊聊怎么突破自己的瓶颈吗？因为你那里还有很多其他技巧，但我只是想反思一下，这真的很有趣：你从第一性原理出发，学习如何以产品经理、工程师或设计师的身份来构建产品，你正在摸索一种工作流程，让AI帮你填补作为工程师、产品经理时所缺乏的技能，帮助你明确需求和设计。

By the way, want to talk about like how you unblock yourself because there's a whole other series of tips you have there, but I just want to reflect on it's so interesting how one you're kind of from first principles, under learning how to build product as a PM, as an engineer, as a designer, and you're kind of figuring out a workflow where AI is helping fill in all the gaps that you don't have for as an engineer, as a PM, helping you craft your needs and design.

所以我觉得这非常有趣。

So I think that's so interesting.

有趣的是，这些功能仍然有效且必不可少。

Just this it's interesting that these functions still work and are necessary.

现在是你和AI共同创造这一切，基本上形成了一个一直存在的三角组合：产品经理、工程和设计。

Now it's you and AI help create all this basically, this triad that's always existed, product manager, engineering, and design.

我一直以来都在思考一个问题：在未来，哪种背景会最有价值？

And something I've always thought is that there's this question of which background will be most valuable in this future.

是产品经理吗？

Is it a PM?

是工程师吗？

Is it an engineer?

是设计师吗？

Is it a designer?

我一直以来都认为，产品经理的职责是澄清、明确要构建什么，清楚地定义需求，弄清楚成功的样子。

My mind has always been the PM function is like their job is clarify, figure out what to build, clarify what to build, be really clear about the requirements, figure out what success looks like.

我觉得这种能力是最需要的。

It feels like that's where the skill is most needed.

此外，设计方面还包括让产品看起来很棒。

There's also a design component of like, make this look awesome.

我觉得，真正擅长设计、品味和判断的价值将会不断提升。

And I feel like that's going to be an emerging that the value of that being really good at design and taste and judgment is only going to go up.

在我们讨论你如何突破瓶颈之前，因为很多时候，事情会偏离方向，出现bug，不谈泛泛而谈，你该怎么办？

Before we get to things you've learned about unblock yourself, because a lot of times, you know, things don't go in the right direction, there's a bug, without being in the generic, what do do?

在进入那个话题之前，你还有什么想分享的关于如何取得成功的建议吗？

Before we get there, is there anything else you wanted to share around just like tips for being successful?

如果我们用正确的方式来衡量成功，正如你所说，AI无论你的背景如何，都是一种放大器。

If we measure success in the right terms, again, AI, as you pointed out, regardless of your background is an amplifier.

所以，如果你不知道自己在做什么，你只是会更快地产出垃圾。

So, you know, if if you don't know what you're doing, you're just gonna produce garbage faster.

有一点，我想再强调一下，在过去，差不多就行了。

One thing, again, I just wanna double down on is in the old world, good enough was good enough.

对吧？

Right?

因为即使产出差不多的东西也不容易。

Because even producing good enough was not easy.

对吧？

Right?

十年前、十五年前，只要能做出来就已经绰绰有余，甚至远超及格线了。

Ten years, fifteen years ago, just producing was more than plenty, more than good enough.

你做了一个SaaS产品，谁在乎它长什么样？

You built a SaaS, who cares how it looks like?

它能运行、能完成任务，天啊，我的效率提高太多了。

It works, it does stuff, like, oh my god, I'm so much more productive.

今天，如果‘还行’就在这里，我们来给人们做个直观的比喻。

Today, like, if good enough was here, let's say let's visualize it for people.

比如，如果这是糟糕、还能更好、平庸、还行、世界级之间的差距，那你觉得呢？

Like, if this was like pretty bad, could be better, mediocre, good enough, world class If this was the gap between good enough and world class, well guess what?

现在，这个差距变成了这样。

The gap is now this.

因为每个人都在用AI生产‘还行’的东西。

Because everybody produces good enough with AI.

绝对每个人都在这么做。

Absolutely everyone does it.

所以，如今最重要的教训是：学习并优化如何产出世界级的、惊艳的作品。

So now learning and optimizing for how do I produce world class and magic is the key lesson to take away today.

正如你所指出的，我认为产品经理是当今AI的最大受益者，因为他们带来了清晰度。

As you pointed out, I think PMs are the winners of AI today because they bring clarity.

如果我要打赌的话，正如人们常说的，我会押注下一波赢家是设计师。

If I was a betting man, as they say, I'd bet that the next class that wins are designers.

因为我们正在训练这些工具，让它们更清晰、更优秀，做出更好的技术决策。

Because we're training these tools to be more clear, to be better, to make better technical decisions.

我认为我们还不会训练它们来做出更好的情感决策。

I don't think we will train them just yet to be make better emotional decisions.

我认为设计本质上是关于情感的。

And I think design is all about emotion.

而这就是需要提升技能的地方。

And that's where, like, the level up the skill up needs to come.

如果问我，我认为这是最大的提升——当你加入Lovable时，你最需要明白的核心是什么？

That's the biggest level up if you ask me, like, oh, what is the main thing you figure out when you join Lovable?

也就是说，你个人最大的技能提升是什么？

Like, what's the biggest personal upskill?

比如说，和Felix、Nad、Abby这些设计师一起工作，才是真正让我有所突破的关键。

Let's say, like, working with Felix, Nad, Abby, all of the people that are designers just really what moved and shifted the needle for me.

我意识到，原来这就是世界级水准，原来需要做到这种程度。

I'm like, oh, so this is how world class looks like and this is what it takes.

对吧？

Right?

我总是用这样一个比喻：我想偷一个他们的设计，用到我的Lovable项目里。

I I always use the analogy of, like, I I wanted to steal one of their designs and bring it into my lovable project.

于是我打开Figma，心想：就拿这个背景，直接放进去好了。

So I went into Figma, I was like, let me just take this background, like, and just put it in there.

我进去后才发现，本以为是个很简单的渐变，竟然用了50个图层才做出来。

I went in and realized that what could be, you know, as a pretty simple or rather simple gradient took 50 different layers to produce.

于是我点开了那个组件。

So I clicked on that component.

我心想：天哪。

I was like, oh my god.

这根本不是三种颜色。

This is not three colors.

这是五十种颜色。

This is 50 colors.