与Maggie Appleton一起探索设计工程

你就能得到那种……

You you get the kind of...

先把方向性的东西搞对，等它变得很棒了，你可能就会去构建它，或者决定去构建它。

Get the directional things right and then once it's great, you might build it or you might decide to build it.

但我肯定……

But I definitely...

我的意思是，希望我们之后能展示这些。

I mean, we can hopefully show these later.

我确实做了很多草图绘制的工作。

I do I do a lot of stuff sketching.

我的意思是，即使是界面设计，很多时候你就是在画一些方框，然后想着，这个从前面滑过来，那个从底部滑上来，比如我点这个按钮。

I mean, even just interface design, a lot of what you're doing is just drawing boxes and then being like, there's this slide over from the front, from the bottom, like, I click this button.

所以这就像你在画UI草图。

So this is like you kind of drawing up UI.

一个你觉得可能看起来像那样的界面。

A How an interface you think could look like.

所以我看到的是，嗯，混合了各种界面。

So so I see I see a mix of Yeah.

有各种界面，我们会把它们放到屏幕上，但这些界面描述了它们的功能，比如，你能说说其中一个吗？或者选一个你觉得有趣或难忘的？

Of mix of UIs and we'll put these on onto the screen, but UIs, description of of what they do, what, like, can you just talk to one of these or or like one that that's interesting or memorable?

这是一个新的……

So this is a new...

我应该先解释一下背景。

I should explain the context.

我现在正在做一个原型，我的想法是，目前智能体的一个瓶颈是规划体验非常糟糕。

I'm prototyping at the moment something where my theory is like, one of the bottlenecks with agents is planning is a really bad experience at the moment.

比如现在，跟一个智能体长时间对话。

Like at the moment, have a long chat with an agent.

你……

You...

有时候甚至在命令行里，那是一个非常原始的界面。

Sometimes even in a CLI, which is a pretty primitive interface.

然后这个代理会通过问你一系列选择题来拷问你，比如A、B、C选项。

And then the agent grills you by asking you a set of like choice a, b, or c questions.

如果你用Matt Pocock的grill me技能之类的，它会这样重复问你好几百遍。

It does this, like, a 100 times over if you're using Matt Pocock's grill me skill or whatever

你喜欢就好。

you like.

他他他刚上过播客。

He he he was just on the podcast.

哦，那正好。

Oh, there go.

到第20个问题的时候，你就已经很累了，大脑开始罢工。

By question 20, you're, like, quite tired and your brain starts shutting down.

我也有同感。

That was my experience.

我坚持下来了，回答了36个问题，然后我就有点……

I went through and I got 36 questions and I I was kinda...

那些问题还行，但大概到20或25个的时候我就开始烦了。

They were good, I was starting to get annoyed around like, yeah, like 20 or 25.

我当时就想

I'm like

而且

And also

而且我完全不知道什么时候会结束。

And I had no idea when it would stop.

没完没了的。

It's endless.

而且人脑就是这样，你会累。

And like the human brain, you get tired.

在这么短的时间里，你没法做出同样的决定。

You can't make the same decisions in this shorter time.

而且大多数决定你都没有足够的信息，因为它给你一个问题三个选项，还告诉你推荐选A。

And also you don't have enough information about most of those decisions because it's given you a question and three options and it's told you number A is recommended.

然后你就开始这样，对，A，A，回车，A，我同意你。

Then you just start being like, yep, A, A, enter, A, I agree with you.

所以这不是一个理想的体验。

So this is not an ideal experience.

这就涉及到，智能体喜欢输出大段大段的文字，这对智能体来说是理想的输出，但对人类来说却不是理想的输入。

This gets into like agents love to output reams and reams of text, and that's an ideal output for agents, but it is not an ideal input for humans.

所以我们存在一个不匹配的问题，人类需要能够消化大量信息、真正理解并做出明智决策，而这并不是适合这种需求的界面。

So we have a mismatch with like, what do humans need to be able to digest large amounts of information and truly understand it and be able to make informed decisions, that is not the interface for this.

所以我在探索如何设计一个界面，让我们能更好地规划和决策，但方式要更容易理解。

So I'm trying to explore how would we make an interface that got us to do better planning and better decision making, but in a way that was easier for us to comprehend.

这就是我目前正在做的原型。

So this is what I'm prototyping at the moment.

其中一部分是，好吧，这将是多人参与的，因为当然一切都要和你的团队一起规划。

And part of this is like, okay, it's gonna be multiplayer because of course everything should be with your team planning together.

所以几个人可以一起做。

So several people can do it together.

对。

Exactly.

但关键是怎么让每个决策有更多的空间？

But also just like how do you make it so that you have more space for each decision?

所以我正在尝试做一个原型，比如，好。

So I'm trying to prototype like, okay.

首先，这肯定得在图形界面里做，不能是命令行。

So first of all, this has to definitely happen in a GUI, not in a CLI.

所以这些只是你关于如何提供更多空间的粗略想法，或者……

So these are just like rough ideas of how you could have more space or...

对。

Exactly.

我在想，每个决策可能都有自己的小文档或卡片。

So I'm thinking of it as like each decision maybe has its own little mini document or card.

而且根据决策的不同，有些可能只需要三个选项。

And depending on the decision, some of them maybe you can just have three multiple choices.

那没问题。

That's fine.

可能是个简单的决定。

It's maybe a simple decision.

选A，或者C。

A, or C.

但有时候它会问我，比如，你觉得边框应该用灰色10%、12%还是15%？

But sometimes it'll ask me things like, do you think the border should be like gray 10%, 12%, or 15%?

我就会说，那你给我看看效果。

And I'm like, well, show show me.

这是个视觉问题。

This is a visual question.

有时候它又会问，你希望架构怎么组织？

Well, sometimes it's like, how do you want the architecture to be structured?

我就会说，那你给我看个架构图。

And I'm like, well, show me an architecture diagram.

给我看一个数据流图。

Show me a data flow diagram.

给我看一个状态机。

Show me a state machine.

就像，我在尝试做原型，比如那些带有图表、原型和嵌入的HTML的决策，这样根据被问的问题，它会显示正确的界面来帮助我做决定。

Like, I'm trying to prototype, like, decisions that have come with diagrams and prototypes and HTML embedded in them so that depending on what question is being asked, it shows me the correct interface to make that decision.

所以我在这里说的原型是指，比如你有一个计划，那是一份大文档。

So what I meant prototyping here is like, okay, you've got a plan that's a big document.

你可能在里面嵌入了一些小的决策文档，这些文档可以……

You've maybe got these little decision docs embedded within them that can...

你可以展开查看更多。

You could expand to see more of.

然后，我觉得每个决策都应该分配一个人，由那个人做出决定，这样以后就有审计追踪了。

And then looking at like, I think each decision should have a human assigned to it who made that decision so that you have an audit trail later.

不一定要追责，而是为了搞清楚，比如，我们为什么在后台做了这个决定？

Not necessarily to hunt people down, but to be like, okay, why why did we make this decision about the back end?

好，那我们去看看吧。

Well, let's go see.

好的。

Okay.

卢克大概六个月前做的那个决定。

Luke made that decision, like, six months ago.

我们去打开他的决策卡片，看看他当时有哪些信息可以依据来做这个决定。

Let's go open up his decision card and be like, what information did he have available in order to make this decision?

希望这以后能派上用场。

Hopefully, this becomes useful later.

不过现在我只是在画草图。

But here, I'm just sketching.

这个可能有哪些不同的形式呢？

What are the possible shapes this could take?

是像一叠卡片那样在屏幕上展开吗？

Is this like a stack of cards that expands on the screen?

这是不是就是一个大的线性流程？

Is this like all one big linear thing?

我们是不是像刷Tinder一样滑动浏览？

Are we like swiping through stuff like it's Tinder?

什么样的界面适合这个？

Like what interfaces are gonna be useful for this?

所以很多草图就是在尝试弄清楚各种可能的形态。

So this is what a lot of the sketching is just trying to figure out like possible shapes of things.

我有很多种。

I have all kinds.

哦，这些是角色，因为我当时想做一个ESP角色。

Oh, these are like characters because I was trying to make an ESP character.

开始吧。

Go.

你能给我看看角色吗？

Can you show me character?

嗯，这个就像……

Well, this is like...

你知道吗……

Do you know...

哦，这就像我孩子们在我的笔记本上乱涂乱画。

Oh, this is like my kids scribbling all over my notebook.

这就像，你知道的，ESP 32 那些东西。

This is like, you know, ESP 30 two's.

你知道 Steve Ruiz 已经说服大家去买 ESP 32 了吗？

Do know Steve Ruiz has like sold everyone on buying an ESP 32?

就是那种带屏幕的小设备，里面有个很小的芯片，带 Wi-Fi。所以我设计了一个小角色，它会告诉我天气，我还想把它连接到我的代理上，这样它就能提醒我，比如“你的预设额度快用完了，还剩这么多”。

It's like this little device with a screen and a very small microchip that has Wi Fi and And so I was designing this little character who like tells me the weather and like, I'm trying to hook it up to my agent so it can be like, you have this much capacity left before your preset limit happens.

所以，这就是各种……

So like, this is just all kinds of

事情在发生。

things happening.

这里面既有严肃的内容，也有一些好玩的东西。

It's a mix of, like like, some serious stuff and some fun stuff.

我还可以展示整个用笔记本和纸上思考的过程。

And I can show, like, the whole thing with notebooks and thinking on paper.

我觉得所有设计师都会这么做，但对我来说，因为我是从插画转过来的，我这么做的原因是我一开始是在一个必须把每样东西都画出来才能想清楚要做什么的世界里。

I think all designers do this, but coming from illustration, the reason I do this is because I started in a world where you have to draw everything to like figure out what you're gonna do.

所以当我做插画师的时候，我有好多这样的笔记本，用来琢磨构图啊什么的。

So when I was an illustrator, I had tons of these notebooks where you're figuring out like, what's the composition?

东西的物理形态是什么样的？

What's the physical shape of things?

比如，你怎么去处理场景里草的形态？

Like, how are you working out like the shape of the grass in a scene?

所以我觉得我很早就开始习惯在纸上解决问题，比如构思构图和布局。

And so I think I started problem solving on paper, like very early on in my career, like figuring out compositions and layouts.

这是大学时候的事吗？

Was this in college?

那是什么时候的事？

When was this?

我记得我是在为Egg Head工作的时候做的这些。

I think I was working for Egg Head when I was doing this.

这些都是你画的吗？

And you drew all of these?

我受过训练，跟……

I was trained with...

我在洛杉矶待过一阵子，那地方真不适合住人。

I was in LA for a while, which is a terrible place to live.

但是跟……

But with...

我跟着一些电影概念艺术家学习过。

I trained with, like, people who are concept artists on films.

他们的工作方式非常漂亮，而且非常技术化。

And they have a really beautiful way of working that's very technical.

这很像用三维形状来构建东西，就像在空间里画画一样。

It's very like constructing things from three d shapes and like drawing in space.

这些真的很立体。

These are really three d.

然后……

And then...

他们基本上就是教你画风景和构图。

And they kinda just teach you how to do like landscapes and layouts.

我很喜欢他们的教学方式。

I loved their way of teaching.

而且，再说一次，这非常技术化。

And it's very, again, it's very technical.

这很像工程学。

It's very like engineering.

你得懂机器人学，比如不同类型的关节，这样你才能把机器人组装起来，画出看起来可信的机器人。

You have to understand like with robotics, like different types of joints you could like put together a robot in so that you could actually draw a robot that was believable.

当时真的需要很多对现实的理解，才能把现实画得让人信服。

There was a lot of like understanding reality in order to believably draw a reality.

所以我觉得这段经历为我后来转做UI设计打下了基础，因为很多这种草图绘制都用得上，你知道的……

So I think doing this set me up to when I moved into, like, UI design to do a lot of this kind of sketching because, you know...

就是自然而然就上手了。

It just comes naturally.

我是说，这些画起来也稍微容易一些，从……

I mean, these are a little bit easier to sketch as well in terms

对，从……

of the Yeah.

用户界面，但是……

User But it...

然后我觉得就像……

Then I think it's like...

就是能帮你在纸上思考，所有这一切就像是你脑子里有个想法。

That it helps you think on paper that all of this is just sort of like you have an idea in your head.

当然，我也可以打开Claude Code，跟它说，嘿，Claude。

And sure, I could, like, go into Claude Code and be like, hey, Claude.

我有个想法。

Here's an idea I have.

就像一叠卡片，还是个手风琴式的。

It's like a stack of cards and it's an accordion.

但更快的方式就是拿支笔或铅笔，就在我旁边的桌上，直接用手画出来。

But it's much faster to just get a piece of pen, like, pen or pencil on the desk next to me and draw that with my hands.

这样省力多了。

It's like way fewer effort.

所以我觉得网上的人老是在说，哎呀，大家以后都用提示词来创作了。

So I feel like people online keep being like, oh, like, everyone's just gonna prompt everything to create.

但在早期阶段，你需要的是快速反馈和很随意的表达，先把东西的大概形状定下来，然后才能用语言告诉AI你想让它做什么。

But but you need something that is, like, quick feedback and very loose in the early stages to figure out the shape of something before you can put into words what you want an agent to do.

而且，这还是个视觉上的东西。

And also, it's visual.

它不是文字。

It's not text.

用手画是不是更有满足感？

And is it is it not a bit more satisfying doing it with your hands?

好太多了。

So much better.

而且你可以看着它。

And then you can look at it.

它不会从屏幕上消失。

It doesn't go away on your screen.

它可以放在你的桌子上，第二天你看到它就会想，哦，对。

It can sit on your desk and you can, like, the next day be like, oh, yes.

我记得我当时在琢磨这个功能应该是什么样子。

I remember I was trying to, like, figure out what shape this this feature should be.

或者你可以画数据图表，想画什么就画什么。

Or, like, you can draw, like, data diagrams, whatever you want.

就是……

It's...

不一定非得是视觉的，但它就是一种把想法外化的方式。

It doesn't have to be visual, but it's just a way of externalizing thoughts.

在它们还没……

Before they're...

那时候它们还不是语言。

Then they're not linguistic.

我觉得关键的一点是，再说一次，智能体只接受文本作为输入。

I think this is a key thing is like, again, agents only accept text as inputs.

对。

Okay.

它们能读图片。

They can read images.

比如我确实会拍照然后放进去。

Like I do take photos and put them in.

但它们在图像方面不太擅长。

But they're not as good at images.

它们在空间推理方面非常差。

They're very bad at special spatial reasoning.

它们在视觉设计方面非常差。

They're very bad at visual design.

我的意思是，让它们做设计，它们总是出错，比如不添加间距，尺寸不对，还会让文字重叠。

I mean, trying to get them to do design, they just make mistakes where like they just don't put spacing around things and things of the wrong size and they make text overlap.

它们看不见，对吧？

They can't see, right?

试图用文字向智能体解释视觉想法真的很难，效果也不太好。

Like trying to explain a visual idea in text to an agent is really challenging and doesn't work very well.

所以我发现，我仍然会在前期自己完成很多设计，因为那只是思考视觉部分。

So I find I still end up doing a lot of my design without agents upfront because all it is is thinking through the visual pieces of it.

然后当我觉得，好的，我知道我想要的东西的形状了，我就可以让智能体去做。

And then when I'm like, okay, I know this is the shape of the thing I want, then I can tell an agent to do it.

但我发现很难在更早的阶段就让它参与进来，虽然我觉得推特上有些人声称他们能做到，但我对此持怀疑态度。

But I find it hard to, like, involve it earlier in a way that I think some people on Twitter are claiming they do, but I'm skeptical.

但这很有意思，因为即使在软件设计中，我指的是架构设计，我观察到的一些最有成效的讨论往往是在白板前面对面进行的。

But but this is interesting because even in software design, I'm talking about architecture design, some of the, I guess, most productive sessions I've observed have been in person around the whiteboard.

在那里，人们，我们又在讨论组件、数据库、网络连接、重试逻辑等等，你知道，这些东西你可以描述出来，或者放到电脑上，但当你把它们画在白板上，当有人画上去，然后另一个人拿过来，他们就被迫把自己的想法放到二维空间里，因为我们不做三维的。

Where people, again, we're talking about like components, databases, networking connections, retry logics, whatever, you know, like these things you could describe or you could put it on a computer, but when you put it on a board, when someone puts it on there and then someone else takes, and they kind of, they're forced to take their ideas into two d space, because we we don't do three d.

你知道，你可以画很酷的三维图，但我们也可以只用方框和箭头。

Know you can draw, like, cool three d, but we can also do just boxes and arrows.

有人画出来，其他人理解了，然后他们加入自己的东西，或者圈出来，或者添加新的组件。

Someone does that, and the other people understand, and they go in and they add their own thing, or they circle, or they add new components.

即使在数字领域，我觉得Miro就是一个很好的例子，它们变得如此流行，是因为它们找到了一种协作白板的方式，你知道，你不需要在同一个房间，但它们提供了类似工具。

And even in digital space, I think Meijer was a good example, they became so popular because they figured out a way to do collaborative whiteboarding for, you know, like, you don't have to be in the same room, but they give you somewhat similar tools.

所以我在想，把想法放到物理媒介上，对我们工程师来说就是白板，对你来说就是笔记本，也许这真的有助于重新思考或巩固想法，而且它确实有一种强制功能，关于动词、形状，你……

So I wonder if if this whole thing of of getting your ideas either to a to a physical medium, which I think for us engineers as the whiteboard, for for you as the the notebook, maybe maybe it just helps maybe rethink or solidify or also it it does it does have forcing function about the the verbs, the the shapes that you...

有意思。

Interesting.

这就涉及到我们现在和智能体之间的交互界面，实在是太原始了。

And this gets into kinda like the interfaces we have to agents right now are so primitive.

我觉得我们都清楚这一点。

I think we know this.

我们都已经进入这个领域好几年了，真是不可思议。

We're all like a couple years into this entire thing, is is wild.

也许五年了。

Maybe five years.

我忘了GPT-4是什么时候发布的。

I forget when GPT-four came out.

是三年前还是四年前？

Was it three or four years ago?

ChatGPT 3.5是三年前发布的。

ChatGPT 3.5 came out in the three years.

是2020年4月吗？

Was it 04/20

或者是22年4月。

or 04/22.

那是我在Elicit的时候。

This is when I was at Elicit.

当时我们还在讨论要不要做聊天界面，然后他们就推出了，我们当时就想，哦，他们好像偷了我们的想法。

Then because we were debating doing a chat interface and then they did it and we were like, oh, they like stole our idea.

偷了我们的……其实也不一定非要跟他们竞争。

Stole our It's best necessarily competing with them.

但那其实也没过多久。

But that is no time at all.

想想也挺神奇的，我们已经走了这么远，但我总是说，软件设计其实是一个非常年轻的领域。

And it's kinda wild we've made it as far as we have, but it's also like, I always say software design is an extremely young field.

我们最多也就搞了六十年吧？

Like what are we sixty years into it at most?

所以即使到现在，我们还有很多关于如何设计最佳界面的事情没搞清楚，让人和机器能更好地沟通，而智能体这块更是几乎没怎么开始。

So even that we haven't figured out a lot of things about how to design the best interfaces for people and machines to like communicate with each other and agents barely at all.

我觉得智能体生活在一个自己的世界里。

I feel like there's this world that agents live in.

那里有各种权重、模型、技能和MCP。

There's like weights and models and skills and MCPs.

而人类这边呢，有物理性、质感、光线、材料这些智能体完全不懂的东西。

And then you have your human side that is physicality and texture and light and materials and all these things agents don't understand.

要找到能让我们在中间相遇、一起创造东西的媒介，这才是真正的难题。

And trying to find artifacts that allow us to meet in the middle and create stuff together is the really hard challenge.

因为这是两种完全不同的存在。

Because you've got two totally different types of beings.

不是说智能体是有意识的生命。

Not that agents are conscious beings.

我可不这么认为。

I'm not in that camp.

但它们确实是一种智能，它们思考和行动的方式跟人类不一样。

But they're a type of intelligence that like wants to think and act in a certain way that is not the way humans want to think and act.

而且这两者之间很难转换。

And it's hard to like translate between the two.

我就是觉得很沮丧，它们不能在我旁边看着我，看我的笔记本，理解我在画什么，帮我推进我的想法。

I just find myself very frustrated that they can't sort of be looking over my shoulder, looking at my notebook and be understanding what I'm drawing and helping me move my ideas along.

这就像最终的梦想，它们能理解空间、光线、形状和线条，还有……

This is like the eventual dream is they understand space and light and shape and lines and and...

你懂吗？

You know?

但我觉得我们离那还很远。

But I think we're quite a ways away from that.

那么有一件事……

So one thing...

我们聊过你的笔记本和数字工具，但你最近发帖提到的是木工。

We've talked about your notebooks and digital tools, but one thing that you've you've posted about recently is woodworking.

你说你……

You said you said you're...

我现在正处于软件设计旅程的这个阶段。

I'm at the stage of my software design journey.

嗯，不是。

Well, no.

我正处于那个开始上木工课的软件设计阶段。

I'm at that stage of software design where you start taking woodworking courses.

你能跟我聊聊那段经历吗？

Can can you can you tell me a little bit about that experience?

这个笑话的梗在于，我觉得每个工程师到了某个时候，都会面临一些选择，因为你会觉得自己跟软件工作的世界脱节了。

Well, the little bit of the joke is like, I just feel like at some point, every engineer, you have a choice of things you can get into because you you feel like you're like disembodied from the world of working in software.

所以你得选陶艺、烤面包。

So you have to pick ceramics, bread baking.

你也许可以选摩托车修理之类的，但木工挺受欢迎的，因为它有点像工程学，有很多测量和精确的要求。

You can maybe pick like motorcycle or repair or something, but woodworking is a pretty popular one because it's like engineering y, there's lots of like measurement and being precise.

所以，是的，我在学木工，因为我在伦敦买了房子，而这里的房子都很老旧。

So, yeah, I'm learning woodworking because I bought a house here and all the houses in London are very old.

所以你买的房子需要很多修缮工作。

So you buy a house that needs a lot of work.

这总是会发生的事。

That is like always what happens.

我想阿姆斯特丹可能也一样。

Maybe the same in Amsterdam, think.

老房子，然后你突然就变成，哦，我需要学点DIY技能，我以前挺怕这个的。

Old So housing then you suddenly become like, oh, I need to learn like DIY skills, which I was scared.

你知道，Claude、ChatGPT，还有YouTube和Instagram，在这方面都是很有用的帮手。

You know, Claude, ChatGPT, very helpful coaches in this regard along with YouTube and Instagram.

我能学到很多东西。

I can learn a lot.

但我就是觉得，哦，有太多东西需要解决方案，我得想办法自己做出来。

But I just was like, oh, there's so many things I need solutions to that I need to figure out how to make.

我需要在这里做架子。

I need to like make shelves here.

我还得修这个扶手。

I need to like fix this banister.

所以我就想，我没什么动手能力，也不会木工，但还是报了个课，想着，好吧，我去学。

So I was like, well, I have no DIY skills, no word working, but just signed up for a course being like, well, I'll learn.

我挺确定我能……

I'm pretty sure I can like...

我还没老到那个程度。

I'm not too old.

我能学新技能，而且我肯定能找到很多和软件相关的共通点，但做出来的东西能亲手摸到，那种感觉更满足。

I can acquire new skills and I'm sure I'll find lots of parallels to software, but in a more satisfying way where you actually like touch the thing you make.

你之前提到过一件事……

One thing that you've talked about...

你一直在说的一个概念是设计工程师。

You've been talking about is the concept of of design engineers.

几年前，你发过……

For for a few years ago, you posted...

我会引用你的话。

I'll I'll I'll quote you.

我有个强烈的...

I'm having a strong...

我是不是应该直接成为一个全职的设计工程师？

Should I just become a full blown design engineer day?

我甚至不确定那是什么意思，但我就是想去写很多代码，在屏幕上解决实际的问题，做出漂亮的动画效果。

I'm not even sure what it means, but I just want to touch a lot of code and solve tangible problems on screens and make beautiful animated stuff.

然后你继续去收集你认识的人的名字，你觉得他们是在做这种设计工程的人。

And then you went over and you tried to collect names of people who you knew, who you felt were kind of doing this design engineering.

那是几年前的事了。

Now that was a few years ago.

关于设计工程师，你现在搞清楚了什么？

What have you figured out about design engineers?

如果他们存在的话，他们是做什么的，在哪里...

If they exist, what they do, where...

他们在哪些地方工作？

What places they work in?

我觉得他们确实存在。

I think they do exist.

我觉得Twitter可能有不同的定义，或者我觉得有很多人不是那样……

I think Twitter might have a different definition or like I think there's a lot of people where not that...

我应该叫它X。

I should call it x.

抱歉。

Sorry.

X。

X.

Twitter X。

Twitter x.

都一样。

All the same.

在X上，我觉得很多被称为设计工程师或者自称设计工程师的人，其实是非常优秀的微交互设计师。

On x, I feel like a lot of people who get called design engineers or who present themselves as design engineers actually are, like, very good micro interaction designers.

比如，像这样，按钮上有个很酷的悬停效果。

Like, little, like, here's a cool hover effect on a button.

还有，像这样，一个很酷的加载过渡状态。

And like, here's a cool loading transition state.

这些东西确实很酷。

And those things are definitely cool.

但我不认为那是设计工程，因为你只要让一个代理去做这些事，甚至不用看一行代码就能实现。

I don't consider that design engineering because you could achieve most of those things by just telling an agent to do those without looking at a single piece of code.

对我来说，那只是非常高级的动效设计和视觉设计。

To me, that's just like very advanced sophisticated motion design and visual design.

那很酷，但我不认为那是设计工程，因为我觉得那不足以成为你整个职业生涯的工作。

And that's cool, but I don't consider that design engineering because I don't think that's like a full job for you to do that as a career.

但现在的设计工程……

But design engineering now...

我认为那些优秀的设计工程师，做的正是我之前说的那种工作，就是更深入地参与到工程那一侧。

The people who I think of as good design engineers do the kind of thing I was talking about before where you step much more into the engineering side of work.

所以你仍然是一名设计师。

So you're still a designer.

你仍然关心产品的名词、动词和视觉设计，但你会真正地和工程师紧密合作，或者……

You're still caring about product nouns and verbs and the visual design, but then you really work with engineers closely or...

或者直接参与实现，自己动手写代码。

And or are like directly involved in implementing and writing code yourself.

而且你完全理解，或者说，不是完全，你不需要是全栈。

And you fully understand or to, know, not fully, you don't have to be like full, full stack.

但你对所构建产品的技术架构有深入的理解。

But like you have a deep understanding of the technical the technical architecture of the product you're building.

你真的会想，好，基于后端数据的结构，界面上能实现什么？

You really are like, okay, given the shape of the back end data, what is possible in the interface?

就是那种设计工作。

Like that sort of design work.

你知道吗？

You know?

根据模型的能力和那种技能……

Given given what models are capable of and what kind of skill...

比如我们正在为这个产品构建的自定义技能。

Like custom skills we're building into this product.

我该怎么向用户解释这个产品的功能呢？

Like, how do I explain to users what the capabilities of this product are?

真正深入技术细节，而不是只停留在用户访谈、市场和侧边栏颜色这些层面，而是更关心并更紧密地与工程师合作。

Like truly digging into the technicals and not living in the world of like just user interviews and the market and like what color is the sidebar, but caring much more and working much more closely with the engineers.

而且几乎总是，我觉得很多工作都是你自己来实现的。

And almost always, I feel like implementing a lot of it yourself.

比如我一直自己做前端工作，因为这样更简单。

Like I've always done my own front end work just because it's easier.

而且和我合作的工程师通常都很高兴，因为他们讨厌CSS。

And then the engineers I work with are usually thrilled because they hate CSS.

然后他们就可以去做更有意思、更难的那些工作了。

And then they get to go work on, the more interesting difficult stuff I'd say.

比如同步啊，或者前端那些底层的活儿。

It's like the syncing or like the the the, like, back of the front end work.

你知道，就是那些更偏逻辑的东西，他们能真正投入进去，不用操心什么圆角半径对不对。

You know, like, the more logic stuff, they really get to engage in that and they don't have to worry about, like, is this the right border radius on something?

我觉得他们其实不太在乎这个……

Like, I don't think they...

他们很少在意，但我在意。

They very rarely care, and I do care.

所以这样分工一直挺顺的。

So it's always worked out well to kinda do that.

所以我把它定义为那种深度参与设计工程方面的人。

So I just define it as someone who's, like, deeply into the engineering side of the design.

你过去的职位、现在的职位，还有你认识的设计师朋友，都是什么样的？

What's in your past positions, current positions, through friends who are also designers?

你见过哪些工程和设计协作得很好的例子？

What have you seen great engineering and design collaboration look like?

是和设计师合作，还是和设计工程师合作？

May that be with designers or design engineers?

我觉得作为设计工程师，我一直和工程师们合作得很愉快，因为，再说一次，你做了他们一直不想做的事，而且你消除了设计师和工程师之间大部分的矛盾，这些矛盾我从来没经历过，因为我一直是做前端的。

I mean, do find being a design engineer, I've always had amazing collaborations with engineers because, again, you do the bit they didn't ever wanna do, and you take away what I assume is most of the tension between designers and engineers that I've never experienced because I've always been a front end person.

但我听说过，有人之间会有那种紧张关系，比如有人做了一个完美的Figma文件，但工程师没有完全按照规格实现。

But I've heard of people, you know, there being the tension where you have like someone's made some perfect Figma file and the engineer hasn't implemented it exactly to spec.

是的。

Yes.

我

I've

你经历过。

You've been there.

经历过。

Been there.

显然，这里存在紧张关系，因为有人用某种媒介做了个东西。

Obviously, there's a tension there because someone made something in a There's medium

一个漂亮的渐变，但在移动端，做渐变会严重影响滚动性能。

a nice gradient and it's mobile and doing gradients would absolutely wreck the scroll performance.

而且，这些都是好几年前的事了，当时这还是个问题。

And again, these are like, we're talking about years back when this was a thing.

现在你回头去改，比如改成单一颜色，或者你试图解释你能做什么，要么是性能限制，要么是考虑安卓的低端设备，或者iOS和安卓的差异，而设计师可能更偏向iOS世界。

And now you're going back, like, I change this to a single color, or you try to explain what you can do with your either performance limitations or think about low end devices on Android, or iOS versus Android, where a designer might be more into the iOS world.

再说，好的设计师不会这样，但经常是工程师过来说，好吧，我们这里有限制，不管什么原因。

And again, a good designer would not, but there's oftentimes the engineer would come like, alright, we have constraints here for whatever reason.

我觉得这涉及到，我不想怪设计师，因为我认为这是工具的失败。

I think this gets into, like, I don't wanna blame the designers because I think it's a failure of tools.

因为大家以前用的设计工具，像Sketch和Figma，这些像素级的模拟图，它们和媒介的限制没有任何关系，比如你为iOS、桌面或网页开发时。

Because the design tools that everyone used to use, Sketch and Figma, these like pixel mock ups, they have no relationship to the constraints of the medium, which is like whatever you're building for iOS or desktop or the web.

你得理解，在任何设计角色中，你都是在特定的材料里工作。

Like, you have to understand, like the materials you're building within any design role.

就像桌子设计师不会对橡木在不同环境下的表现一无所知，对吧？

Like a table designer would not like be oblivious to like how Oak performs in certain contexts, right?

或者松木会以某种方式凹陷。

Or like how pine dents in a certain way.

同样，如果你在为网页做设计，却不了解性能，或者你的应用是如何获取数据的，加载时间是多少，万一出现竞态条件怎么办。

And in the same way, think if you're designing for the web and you don't understand like performance or like how is your app like fetching data and what's the loading time and what if there's race conditions.

我觉得如果你对这些一无所知，最终会得到糟糕的设计方案，然后人际关系也会变得很糟糕。

Like, I think if you're oblivious to that, you'll end up with bad design solutions and then with really bad relationships.

所以我希望智能体真的能帮助解决这个问题，对吧？

So I'm hoping agents actually help solve this, right?

因为那样的话，设计师使用智能体就能更多地涉足工程领域。

Because really then designers using agents can step more into the engineering side.

同时也可以把智能体当作教练和学习工具，比如说，工程师回来说我们遇到了一个我从来没听说过的错误状态。

And also use agents as a coach and like a learning tool and say, okay, the engineers come back and told me that we have this like error state I've never heard of in my life.

比如，给我解释一下为什么会出现这个错误状态，帮我构建一个状态机。

Like, explain to me why that error state would occur, you know, and help me build a state machine.

现在有这么多新工具可以用。

Like, there's, like, all these new tools available.

但我得说，那些老问题其实都是工具层面的。

But those old frictions, I have to assume, were just, like, tool based things.

我觉得有些设计师可能还停留在过去，不愿意放弃做像素级原型图的做法。

And I would say designers that are maybe stuck in the past who don't wanna let go of like, oh, I make pixel mock ups.

我不知道现在还有没有人这样。

I don't know if anyone's there anymore.

但如果还有的话，那你们的关系以后肯定处不好。

But if they still are, I mean, you're just...

我实在想象不出这种关系以后能好到哪去。

I can't imagine that relationship going well in the future.

然后你提到了Figma。

And then you you mentioned Figma.

Figma有一阵子特别火。

Figma was such a popular tool for a while.

那是工程师和设计师之间的接口。

That that was the interface between engineers and designers.

你今天怎么用Figma，或者过去是怎么用的？

How do you use Figma today or or how have you used it in in the past?

我觉得我一直用它，但从来不是为了做高保真模型。

I think I've always used it, like, not to get to high fidelity mock ups at all.

我觉得是为了把东西的大致形状画出来，比笔记本更进一步。

I think to get, like, the rough shape of things past a notebook.

所以笔记本可以表示，好，形状是这样的，但那是很粗糙的草图。

So like a notebook, can be like, okay, the shape of it is like this, but it's such a rough sketch.

我觉得Figma很有用，可以确定，比如，什么颜色对比度合适，能吸引注意力到页面上的这个元素，或者文本需要多大才能排版流畅。

And Figma is useful, I find, for being like, okay, exactly what color is like the right amount of contrast to like draw attention to this element on the page or like exactly what size does this text need to be to like flow well.

但我从来不会把它做到高保真，因为当然一切看起来总是不同的。

But then I would never take it to high fidelity because of course everything always looks different.

我的意思是，我主要做网页设计。

I mean, I primarily designed for the web.

在浏览器里看起来完全不一样，对吧？

Everything looks different in the browser, right?

这取决于字体渲染、各种细节、响应式设计，还有断点设置在哪里。

Depends on the font rendering and all this kind of stuff and responsiveness and then when exactly your breakpoints.

所以我总是很早就把它放到浏览器里。

So I would always take it into browsers pretty early on.

在Figma里做到中等保真度，比如形状，然后放到原型里，再真正去调整和打磨。

You get like medium fidelity in Figma, like the shape of it, and then you take it into a prototype and then you can really tweak and refine.

当然，有了Agents Now，我只要把Figma设计稿指给代理，说，把这些都弄进去。

And of course with Agents Now, it's just like, I point an agent at my Figma mock up and I'm like, just get all of that in there.

然后做一个所谓的“jig”，就是给一些滑块和变量挂上，就像……jig这个词来自木工。

And make, you know, what's called a jig, which is where you get like little sliders and variables attached to like a little- A A A jig comes from woodworking.

或者做一个专门干一件事的小工具。

Or you make like a little device that helps you do one specific job.

所以在设计时，当你在做实时原型，你会说，好，这些是我拿不准的变量。

So with design, when you're working on a live prototype, you say like, okay, here are the variables I'm not sure about.

我不太确定我的标题大小。

I'm not sure about like my headline size.

我不太确定这些颜色。

I'm not sure about these colors.

给我滑块、颜色选择器，还有这些，或者我不太确定的动画曲线。

Give me sliders and color pickers and like all these things or like the animation curve I'm not sure about.

给我……

Give me...

然后我会实时调整。

And then I'll and then I'll tweak it live.

当我在实时版本中把值调好后，我们就把这些值定为最终值。

And then when I have the values just right in the live version, then we'll commit those to be the actual values.

快多了。

Much faster.

就像按需构建你自己的Figma一样。

It's like build your own Figma as needed.

那在GitHub里面，设计师是怎么工作的呢？

And inside of GitHub, how do designers work?

你和其他的设计师是怎么工作的？

How do yourself and other designers work?

因为那是一个比较大的组织。

Because it's a bigger organization.

你们是在为开发者打造一个工具，这个工具有很多不同的部分。

You're building a tool for for developers with bunch of bunch of different parts.

其实，我和GitHub设计团队本身接触的时间并不多。

I mean, I don't spend that much time with, like, the GitHub design org proper.

我认识里面的一些人，但是……

I know some people in there, but...

哦，因为你们是Next团队。

Oh, because you're Next.

所以GitHub Next团队有点独立。

So the GitHub Next team is a little bit isolated.

虽然我们和公司其他部门关系不错，但某种程度上也是有意为之，我们有点像旁边的研发团队，专门搞些稀奇古怪的东西，然后试着说服公司其他人……

Like we still have good relationships with the rest of the org, but a little bit by design, we're supposed to be kind of like the R and D team out on the side kinda doing weird stuff and then trying to convince the rest of the org what like...

我们是对的，他们应该关注我们，采纳我们的做法，这完全是另一套政治博弈。

That we're right and they should like pay attention and do our thing, which which is a whole different politics thing.

所以我其实偶尔会见到那些设计师，但他们……

So I, like, kind of see the designers, but they...

我觉得他们的工作和我很不一样，他们得维护质量标准，还要负责一个大型设计系统，以及一个老旧的Ruby on Rails应用。

I think they have a very different job to me and that, like, they have to, like, uphold quality standards and they're working on, like, a big design system and, like, an old Ruby on Rails app.

他们面临的限制和我完全不同。

Like, they've got very different constraints to me.

而且他们有数百万甚至数千万的客户，这些客户现在做的事情也不一样了。

And they have, like, millions or or or even more, like, tens of millions of customers who, like, does now a different thing.

他们已经习惯了某些东西，或者习惯了找某些东西，或者习惯了某些模式。

They've been used to certain things or finding certain things or certain patterns.

没错。

Exactly.

对。

Exactly.

在 .com 上改任何东西都是一件很政治化的事情，因为客户习惯了，他们进来后有关键的工作流程。

Changing anything on .com is, like, a whole political thing because because customers expect, you know, them to come in and they have critical workflows.

你不能动他们的按钮。

You can't move their button.

但我所在的团队，就我和另一个同事，我们俩都是设计工程师。

But the team I work on, there's me and one other guy who are both design engineers.

所以我们俩都做工程工作，但我们更偏向设计。

So we both, like, do engineering work, but we are more the design people.

其他人都更偏工程，比如……

And everyone else is much more engineer, like...

而且他们都很资深。

And they're all really senior.

跟他们合作很有趣，因为他们什么都见过。

So it's fun to work with them because, like, they've seen everything.

他们很懂怎么架构应用。

They like know how to architect an app.

不过我们和工程师的工作方式差别不大，真的。

But we don't work that differently to engineers, find.

就是说，我们和他们一样提交代码审查。

I mean, we're putting in PRs the same as them.

我们用的是同样的工具和材料。

We're like working on the same tools and materials.

更多是关注应用的哪些部分。

It's much more that thing of like what bits of the app we care about.

我当然关心界面部分，而我的工程师搭档们当然关心后端、数据流这些。

I, of course, care about the interface part, and then, like, my engineer partners will, of course, care about, like, the back end and the data flow and everything.

所以工具一样，但考虑的东西不一样，大概是这样。

So same tools, but different, like, considerations, I guess.

我想花点时间提一下我们的赞助商，turbopuffer。

I wanted to take some time to mention our presenting sponsor, turbopuffer.

Turbo Puffer 作为这期节目的赞助商再合适不过了，因为他们拥有我认为目前科技界最令人耳目一新的品牌之一。

Turbo puffer couldn't be a better sponsor for this episode as they have what I think is one of the most refreshing brands in tech right now.

Turbo Puffer 的联合创始人 Simon 形容他们的品牌是硬核又带点奇思妙想，而他们的设计也完美体现了这一点。

Turbo puffer's cofounder Simon describes their brand as hardcore and whimsical, and their design perfectly reflects this.

TurboBuffer 这个数据库，是极其硬核的 AI 基础设施。

TurboBuffer, the database, is extremely hardcore AI infrastructure.

它支持着 Entropic、Notion、Bridgewater、Cognition 以及全球许多最大、增长最快的 AI 公司的关键搜索工作负载。

It supports critical search workloads for Entropic, Notion, Bridgewater, Cognition, and many more of the largest and fastest growing AI companies in the world.

他们非常重视可靠性、性能和可扩展性。

They take reliability, performance, and scalability very seriously.

但 TurboPuffer 这个团队，拒绝成为又一个死板、无聊的企业科技公司。

But turbopuffer, the team, refuses to be another sterile, boring enterprise tech company.

看看他们的网站就知道了。

Just look at their website.

这正是你期望中数据库网站应该有的样子。

It's exactly what you would want the website for a database to look like.

它简单、低调，而且很透明。

It's simple, modest, and transparent.

他们甚至把限制条件直接放在首页上。

They even put the limits on the homepage.

但同时也一点都不无聊。

But it also isn't boring.

看起来非常舒服。

It's nice to look at.

设计得非常用心。

It's very well thought out.

他们那些小小的ASCII图，把硬核的工程技术和俏皮的设计完美地结合在了一起。

And their little ASCII diagrams perfectly blend hardcore engineering with whimsical design.

而且，如果你试过用AI生成一张像样的ASCII图，你很快就会明白，这些图都是他们手工做的，不是AI生成的。

Plus, if you've ever tried to generate a good ASCII diagram with AI, you'll quickly realize that they create all of these by hand, not with AI.

TurboPuffer的品牌形象完美地反映了他们的本质，这也是它之所以这么成功的原因。

TurboPuffer's brand is the perfect reflection of who they are and that's why it works so well.

这个品牌既硬核又充满奇思妙想，团队也是如此。

The brand is hardcore and whimsical and so is the team.

我强烈推荐turbopuffer，不仅因为它是一个为AI打造的极具扩展性的搜索引擎，更因为他们的团队会竭尽全力为客户服务。

I highly recommend turbopuffer not only as the extremely scalable search engine for AI but also as a group of people that will go to exceptional lengths to deliver for their customers.

你可以在turbopuffer.com/pragmatic上了解他们。

You can check them out at turbopuffer.com/pragmatic.

我还想提一下我们的长期赞助商Entire。

I also want to mention our seasoned sponsor Entire.

Maggie提到，作为设计工程师，她仍然和其他人一样提交PR。

Maggie mentioned that as a design engineer she's still putting in PRs the same as everyone else.

每一个PR都会被推送到Git托管平台上。

And every one of those PRs is pushed to Git hosting.

但git代理存在一个问题。

But there is a problem with git agents.

在现代以代理为中心的软件开发中，Git正日益成为瓶颈。

Git is increasingly becoming a bottleneck for modern agent heavy software development.

开发者们正在用智能体生成更多代码。

Devs are creating more code with agents.

这些智能体在推送更多代码，而且很多开发者还在并行运行多个智能体，推送的代码量就更大了。

These agents are pushing more code, and many devs are running parallel agents that are pushing even more code.

GitHub 明显有点跟不上，还经常出故障。

GitHub is clearly struggling to keep up and has frequent outages.

那解决方案是什么呢？

So what is the solution?

Entire 是由 GitHub 的上一任 CEO Thomas Domke 创立的，他从零开始重新构建了 Git 托管服务，专门为智能体时代设计。

Entire was founded by GitHub's last CEO Thomas Domke and he rebuilt git hosting for the agenda care from scratch.

Entire 的设计目标就是速度极快，并且把你的仓库放在离你近的区域，减少延迟，这样就能让成群的智能体并行推送代码。

Entire was built to be very fast and have your repos regionally close to you to reduce latency allowing for fleets of agents to push in parallel.

而且 Entire 的性能是顶级的。

And Entire's performance is next level.

这个仓库每秒能处理 418 次推送，比市场上所有其他竞争对手快 89 倍。

The repo can handle four eighteen pushes per second and that's up to 89 times faster than every other competitor on the market.

当GitHub宕机时，你仍然可以继续工作。

When GitHub is down, you can still keep working.

而且你不需要从GitHub迁移出去。

And you don't need to migrate away from GitHub.

你只需注册Entire，平台就会镜像你的仓库。

You just sign up to Entire and the platform mirrors your repo.

哦，还有一件事。

Oh, and one more thing.

你有没有想过，这段代码是由什么提示词生成的？

Have you ever wondered what prompt resulted in this code being generated?

我觉得与代理的提示词和对话比PR本身包含更多信息。

I find that the prompt and conversation with the agents carries more information than the PR itself.

Entire会把与代理的所有提示词历史直接记录在仓库里，方便回顾。

Entire captures all the prompt history with your agent right in the repo, easy to check back.

而且它有一个相当创新的界面来展示所有这些。

And it's got a pretty innovative UI to show all of this.

如果你在找即使GitHub挂了也能用的git托管服务，就去entire.io/pragmatic，安装CLI，一键镜像你的仓库。

If you're looking for git hosting that works even when GitHub is down, head to entire.io/pragmatic, install the CLI, and mirror your repo with a click.

我已经这么做了。

I've already done it.

哦，我有没有提过它支持任何智能体，而且是开源的。

Oh, and did I mention that it works with any agent and is open source.

那随着LLM的出现，你的设计流程有什么变化？

And then with LLM, since they came out, how has your design process changed?

你已经提到现在容易多了。

You've already mentioned how it's now a lot easier.

比如，如果你有界面或设计，你可以让Claude或Codex或其他任何智能体来实现它。

You can, like, for example, if if you either have an interface or design, you can you can ask Claude or Codex or or any any other agent to implement it.

但有什么变化呢？

But what has changed?

比如，从笔记本到你的原型，再到最终成果，哪些部分更容易了，哪些部分更棘手？

Like, between the notebook, between your mock ups, between the prototype that that gets gets there, what parts are easier and maybe what parts are trickier?

确实一切都更快了，而且感觉也轻松了不少。

It's definitely all faster, And it definitely all feels a bit easier.

我得说，我确实记得以前在Figma里做那些复杂的原型有多痛苦，因为那比在网页上做更复杂的东西要快。

I will say, like, I definitely remember not suffering through like making elaborate prototypes in Figma because that was gonna be faster than like building more complex things in the web.

然后还得用那些Figma点击式原型给用户看，显然一切都是假的。

And then trying to show them to users with these like Figma click through prototypes that obviously like everything is faked.

哦，我记得那个。

Oh, I remember that.

就像在用户测试的时候，有时候你得搞那些奇怪的东西，点这里然后……

Like on the user thing and then sometimes you have to like have this weird things that click here and then...

所以你在用户访谈中，用户想点某个东西，但你还没把它连接到假的新界面。

So you're like in these user interviews and like the user's trying to click something that you haven't hooked up to like a like a fake new screen.

整个事情感觉有点不真诚。

The And whole thing just feels a bit insincere.

而且即使你做了原型，也没时间把它们做得好看。

And then even when you did make prototypes, couldn't spend any time making them look nice.

所以有时候用户会因为界面看起来很糟糕而有点分心或困惑。

So then sometimes the users would get like a bit distracted or confused by the fact it looked terrible.

但你会说，是啊，但我们只有几天时间，然后就得继续做下一个原型了。

But you're like, yeah, but we only had a few days and then we've gotta move on to the next prototype.

而现在我完全没有这个问题了，对吧？

And now I wouldn't have that problem at all, right?

现在我可以做出看起来相当精致的东西，而且不是靠那些代理在上面操作。

Like now I can build something that really looks quite sophisticated and not with the agents doing it on there.

我发现要做出我认为可以接受的界面，需要很多指导，但速度快多了。

Need a lot of direction I find to make something that I would think was an acceptable interface, but it's so much faster.

有时候我甚至在Figma里做了高保真的模型，然后直接说，嘿，代理，这是Figma文件，你知道的，目标模式。

Sometimes I have even mocked up quite high fidelity stuff in Figma and then just been like, hey, agent, here's the Figma file, you know, goal mode.

你直接做，截图。

You play right, take screenshots.

如果看起来不像Figma里的样子，就继续，不断循环，直到它和Figma完全一致，它最终能做到。

If it doesn't look like the Figma, keep going, keep looping until it looks exactly like the Figma and it'll get there.

所以那就像，我可以让它跑一晚上。

And so that was just like, I can run that overnight.

早上我什么都不用做。

I don't have to do anything in the morning.

界面基本上就按我设计的出来了，就像……

The interface is pretty much there as I spec it out like...

对你来说，这大概就是苦力活吧。

And to you, I guess this was grunt work.

完全同意。

Totally.

确实是。

It's totally.

因为你会觉得，这又不复杂。

Because you're like, this isn't complex.

这就像，好吧，又来了，我们又要做一个侧边栏。

This is like, okay, again, we're building another sidebar.

咱们把React组件弄好，然后……

Let's get the React components and like...

而且你其实不太在乎代码质量或者技术深度这些。

And and you don't really care about the code quality or tech depth or any of that.

反正最后都是要扔掉的。

It's gonna be throw away anyway.

如果它因为那样能跑起来，那就行了，因为之后你会正式做。

If if if if it works because of that, because then you'll build it.

如果跑不起来，那也挺好。

And if it doesn't work, like, great.

无所谓。

Doesn't matter.

所以现在原型制作完全是另一回事了。

So prototyping is a whole different game now.

我觉得你可以做更宏大的原型了。

I feel like you can prototype much more ambitious things.

而且我真的很喜欢整个这种即兴发挥的做法。

And then I do love the whole jig thing.

感觉就像是一种可塑的软件，你可以随心所欲地塑造它，而不受Figma现有工具的限制。

It feels like this malleable software thing where, like, you can make the software whatever you want it to be where I'm not constrained by Figma's available tools.

我可以这样，比如，我最近想做一个动态的星座图。

I can be like, okay, I'm trying to make like this animated constellation map with something I recently did.

比如，星星移动的速度是多少？

Like, what speed are the stars moving at?

比如，我们处于什么缩放级别？

Like, what zoom level are we at?

比如

Like

你做了那个？

You did that?

然后，比如，怎么...

And like, how...

屏幕上的渐变角度是多少度？

What degree is the gradient at on the screen?

然后你把这些都放在小滑块上，调整一下，然后你会想，哦，我真的想要一点绿色的颗粒感覆盖在上面。

And you just put these all on little sliders and you just tweak it, and then you think, oh, I I really want, a slight, like, green, like, grain over this.

告诉智能体。

Tell the agent.

你知道吗？

You know?

就像……

It's like...

说实话，很多时候感觉就像魔法一样。

It honestly feels like magic a lot of days.

就是……

It's...

你有那个吗？

Do do you have that?

哦，我可能能把它调出来。

Oh, I might be able to pull it up.

让我看看。

Let's see.

这是我为新的GitHub Next网站做的东西。

So this is something I built for the new GitHub Next, website.

我们有一个基础版本，但我想要更花哨一点的。

We have a kind of basic one up, but I wanted something a bit fancier.

我们想让发布研究变得更频繁、更容易一些。

We wanna make it a little bit easier to publish research frequently.

所以我一直在玩一些炫酷的主页设计。

So I was playing around with some jazzy homepages.

整个这个东西，就像一个旋转的动画星座。

And this whole thing, this is like an animated spinning constellation.

这些是团队这些年做的一些项目。

And these are like some projects that the team's worked on over the years.

太棒了。

Amazing.

像这种东西，用来做原型，对吧？

And like this kind of thing to prototype this, right?

当然，我是从草图和笔记本开始的，但后来做了一堆小工具，我就想，背景用什么颜色呢？

Of course, I started with the sketch and notebook, but then made a bunch of jigs where I was like, okay, what color is the background?

还有，这个地图转得有多快？

And like, how fast is this map spinning?

然后星星上有很多酷炫的物理效果，但我根本不用做任何物理计算。

And then there's a bunch of like cool physics on the stars that I didn't have to do any physics.

你知道，我完全不需要懂那些。

You know, I didn't have to know anything about that.

你知道，鼠标悬停上去的时候它们有多大？

You know, how big are these when you hover over them?

然后我做了个很酷的滚动效果，直接滚到正式页面，所有这一切，对吧，我一行代码都没写。

And then I did this like cool scroll effect down into the proper page And all of that, right, I didn't write any code.

我做了很多不同的夹具，然后不停地调整和微调，直到完全合适。

I made a bunch of different jigs and and tweaked and adjust exactly

想象一下，这些就是屏幕上的东西。

how that these were, like, just imagine, these were, like, things on the screen.

比如说，颜色从蓝色、绿色、红色这样选，然后你决定，好吧，你最喜欢这个，大小也是，所有这些。

For example, the color you go from, like, blue, green, red, and then you figure out, alright, you like this the most, the sizes, all of those.

没错。

Exactly.

我可以直接说，给我做一个滑块，用来控制星座实验里的动画变量。

I can be like, make me a slider for controlling the animation variables on the constellation experiment.

那就让它先这样吧。

So we'll let that work.

但这就是那种……

But that's the kinda...

其实真的很简单。

It's, like, really simple.

如果你告诉它，比如，给我做控件，或者给我做一个滑块，不管你想做什么，就想想，嗯，我需要怎么调整它才能让它工作？

If you tell it, like, make me controls or make me a slider for whatever it is you're trying to do, just think about, like, okay, what are the ways I need to, like, tweak this to make it work?

它在这方面挺擅长的，直接就能做出来。

It's pretty good about just making them.

我是说，我有这个技能，所以它知道怎么做才能让它们看起来不错。

I mean, I have a a skill, so it, like, knows how to, like, make them so that they look decent.

而且根据你的偏好，你知道的，你摸索出来的那些有效的方法。

And and based on your preferences, you know, what you figured out works.

但这样就很棒了。

But then it's great.

直接连上，然后你就可以随便玩了，这完全是另一种设计方式。

Just hooks it up, and then you can just kinda, like, play with it, and it's it's a whole different way to design.

这非常……

It's very...

你知道，Brett Victor，就是那个做实时编程的。

You know, Brett Victor, like the live programming stuff.