创始人必须亲自掌控招聘、融资、战略和产品愿景

这里是 Naval，你正在收听 Naval 播客。今天我们要讨论招聘、团队和文化。Vinod Khosla 有句名言：你组建的团队就是你打造的公司。换句话说，他们告诉你这是技术游戏，其实这是招聘游戏。所以我翻出了 Naval 在 2025 年 8 月的一条推文：创始人可以委派一切，除了招聘、融资、战略和产品愿景。

This is Naval and you're listening to the Naval podcast. Today we're going to be talking about recruiting, hiring, team, and culture. There's a famous quote from Vinod Khosla: the team you build is the company you build. Or in other words, they told you it was a technology game when it's really a recruiting game. So I pulled up a tweet from Naval from August 2025. Founders can delegate everything except recruiting, fundraising, strategy, and product vision.

招聘是最重要的事，因为你需要创造力，需要积极的人。理想情况下，早期员工都是天才。他们自我管理、低自我、勤奋、能力极强的建设者，技术型，可能有一两个销售，但你不可能事事监督，不可能事无巨细地管理。早期员工是公司的 DNA。当你外包招聘，当别人在没有你直接参与和否决权的情况下进行招聘、面试和决策时，那是悲哀的一天。那是公司不再由你直接驱动的一天。中间出现了一个电传操纵的元素，通过另一个人进行机械连接，往往还很遥远，而其他人不会有你作为创始人的那种筛选标准。

Recruiting is the most important thing because you need creativity. You need motivated people. Ideally, the early people are all geniuses. They're self-managing, low ego, hardworking, highly competent builders, technical, maybe one or two sellers, but you can't watch everything. You can't micromanage everything. The early people are the DNA of the company. When you outsource recruiting, when you have other people hiring and interviewing and making hiring decisions without your direct involvement and veto, that's a sad day. That's the day that the company's no longer being driven directly by you. There's now a fly-by-wire element in between. There's some mechanical linkage going through another human, often at a distance, and other people are not going to have the same level of selectivity that you will as founder.

公司开始变化的重要规模不是某个随意数字，比如 20、30 或 40。而是创始人不再直接招聘和管理每个人的时刻。一旦出现中间管理层，你就与公司有些脱节，直接推动产品团队从 0 到 1 的能力就消失了。所以你真的不能外包招聘。人们以为可以，比如他们雇佣招聘人员。也许你可以外包一点点人才搜寻，但我甚至认为那也很困难。

The important size at which a company starts changing is not some arbitrary number like 20 or 30 or 40. It's the point at which the founder is not directly recruiting and managing everyone. The moment that there are middle layers of management then you are somewhat disconnected from the company and your ability to directly drive a product team that can take the company from 0 to 1 goes away. So you really cannot outsource recruiting. People think you can. They hire recruiters, for example. Maybe you can outsource a little bit of sourcing, but I would even argue that's difficult.

招聘如此重要的原因，很多是显而易见的。我跳过那些明显的原因。但一个不太明显的原因是，最优秀的人真的只想和最优秀的人一起工作。和任何不在他们水平的人一起工作，对他们来说是一种认知负担。他们周围不如他们的人越多，他们就越敏锐地意识到自己属于别处，或者应该做自己的事。最好的团队是相互激励的。他们互相加强。每个人都试图给彼此留下深刻印象。

The reason recruiting is so so so important, and a lot of it is obvious. I'll skip the obvious reasons. But one non-obvious reason is that the best people truly only want to work with the best people. Working with anyone who's not at their level is a cognitive load upon them. And the more people they're surrounded by who are not as good as they are, the more keenly they're aware that they belong somewhere else or they should be doing their own thing. The best teams are mutually motivated. They reinforce each other. Everyone's trying to impress each other.

一个很好的测试是，当你招聘新人时，你应该能对他们说：‘走进那个房间，其他团队成员都在那里，随便挑一个人，随机选，拉出来聊 30 分钟，面试他们。如果你对他们不印象深刻，就不要加入。’当你做这个测试时，你会本能地回避让他们随机面试某个你心里有点数的人。那个人就是你需要放走的，因为正是那个人阻碍你拥有一个每个人都想互相打动的高效团队。这就是你必须保持的标准。

One good test is when you're recruiting a new person, you should be able to say to them, 'Walk into that room where the rest of the team is sitting, take anyone you want, pick them at random, pull them aside for 30 minutes, and interview them. And if you aren't impressed by them, don't join.' When you do that test, you will instinctively flinch at the idea of them interviewing randomly a certain person that's kind of in the back of your mind. That's the person you need to let go because that's the person keeping you from having this high-functioning team that all wants to impress each other. That's the bar you have to keep.

特别是对于你将直接管理的所有人，前 20 个、前 30 个、前 40 个、前 10 个，无论那个数字是多少。在你直接管理的早期员工中，你在寻找什么？有沃伦·巴菲特的老话：智慧、精力、正直。我会加上低自我。低自我的人更容易管理。他们往往较少卷入人际冲突。他们更关心工作，而不是搞政治或争功。你就能更好地规模化。你能管理 30 或 40 个低自我的人，而可能只能管理 5 个高自我的人，因为你总在安抚他们的自我。

Especially for all the people you're going to directly manage, the first 20, the first 30, the first 40, the first 10, whatever that number is. In those early people that you directly manage, what are you looking for? There's the old Warren Buffett line of intelligence, energy, integrity. I would add low ego. Low ego people are just much easier to manage. They tend to engage less in interpersonal conflict. They care more about the work than about politicking or fighting for credit. You just scale better. You'll be able to manage 30 or 40 low ego people when you might only be able to manage five high ego people because you're always massaging their egos.

所以我认为 Vinod 的话完全正确。你组建的团队就是你打造的公司，尤其是你直接管理的前 N 个人。他们是公司的 DNA。

So I think Vinod's phrase is absolutely correct. The team you build is the company you build, especially for the first N people that you are directly managing. They are the DNA of the company.

回到那条推文，你不能外包融资，因为投资者押注的是你。如果你外包融资，那么被你外包的人实际上就是经营公司的人。好的投资者肯定不会支持一个由代理融资的公司，这就是为什么通过银行家融资的公司总是起步就错了。你不需要银行家为你融资。当然，在后几轮中，情况有所不同，因为你接触的是常规风投市场之外的资金，但尤其是在早期阶段，如果你请银行家，那表明存在更深层次的问题。

Back to the tweet, you can't outsource fundraising because investors are betting on you. If you're outsourcing fundraising, whoever you're outsourcing to is really the person running the company. Good investors certainly are not going to back a company where there's a proxy fundraiser, which is why companies that raise money through bankers are always starting off on the wrong foot. You shouldn't need a banker to raise money for you. Now, in later rounds, it's a little different because you're reaching money that's outside of the normal venture market, but especially in early stage, if you're engaging a banker, that's symptomatic of a deeper problem.

战略，你必须制定并传达战略。产品愿景。这是有争议的一点。有些创始人外包产品愿景。但我认为，因为你的工作是把你能找到的最好团队，将他们的能量提炼成完美的产品，将他们的知识和创造力实例化到产品中。你需要统一产品愿景。一个人需要把任何复杂产品完全装进脑子里。这就是为什么有多个创始人会有帮助，因为很少有人能同时融资、招聘并把产品愿景完全装进脑子里。史蒂夫·乔布斯是其中之一。埃隆·马斯克可能也是。但通常你看到的是两人团队：一个更擅长销售，尽管有建设者背景会更好，这样他们知道自己在说什么；另一个更擅长建设，但有一点销售基因也有帮助，因为他们可能要去招聘其他建设者。我认为这四件事你都不能外包。

Strategy, you have to set and communicate the strategy. Product vision. This is the one that's up for debate. There are some founders who outsource product vision. But I would argue that because your job here is to take the best team you can find and distill their energy into a perfect product to instantiate their knowledge and creativity into a product. You need to unify the product vision. One person needs to hold any complex product entirely in their head. And this is where it helps to have more than one founder because it's rare that someone can fundraise and recruit and hold product vision entirely in their head. Steve Jobs was one of these people. Elon Musk is probably another one. But usually you see a two-person team. One person who's better at selling, although it helps if they have some builder background, so they know what they're talking about, and one person who's better at building, but it helps if they have a little bit of a seller bone because they're probably going to be recruiting the other builders. I don't think you can outsource any of those four.

也有产品愿景被外包的案例，下面有某个聪明人在推动产品愿景，但那些很少见。通常，这四件事都由核心创始团队处理。

There are cases where product vision has been outsourced. There's some brilliant person underneath who's driving the product vision, but those are rare. Usually, all four are handled by the core founding team.

公司里永远不会有比创始人更好的招聘者。我这话有两层意思。第一，在任何成功的初创公司中，创始人总是出色的招聘者。但另一面是，创始人作为招聘者、作为人、作为贡献者的质量，是你将带入组织的任何人的质量上限。你永远无法雇佣比你更好的人，对吧？人们说雇佣比你更好的人。我不认为那真的有效。比你更好的人不会长期为你工作。当然，以后当你建立了大企业，有网络效应、有惊人产品时，可能不同，那时你可以雇佣比你更好的人，因为你带来的远不止你自己。但早期，你带给初创公司的只有你自己。要让人们愿意为你工作，你至少得和他们水平相当。这就是为什么我认为早期投资者如此看重创始团队。

There will never be a better recruiter in the company than the founder. And I mean that in two ways. One, in any successful startup, the founder is always a great recruiter. But there's also the flip side of that, which is the quality of the founder as a recruiter, as a human being, or as a contributor is a cap on the quality of anybody you're going to bring into the organization. You're never going to be able to hire anybody who's better than you are, right? People say hire people who are better than you. I don't think that really works. People who are better than you don't want to work for you for long. Now, it may be different down the road when you've built a huge enterprise and there's a network effect, an amazing product, then maybe you can hire people who are better than you because you're bringing a lot more than just you. But early on, all you're bringing to the startup is you. And for people to want to work for you, you have to be at least on their level. This is why I think early stage investors judge the founding team so heavily.

他们甚至不在乎早期进展，至少优秀的人不在乎，也不在乎合作伙伴或领域专长。他们只想看你有多优秀。而展示你有多优秀最清晰的方式，就是招募优秀的人才。

They don't even care about early progress, at least the good ones don't, or about partnerships or domain expertise. They just want to see how good you are. And the clearest way you can show how good you are is by recruiting great people.

你不能外包招聘的另一个原因是，招聘需要极大的创造力。否则，你就会做和其他公司一样的千篇一律的事情，最终得到和其他公司一样可互换的人才。所以，不要把你不懂招聘这件事当成坏事。

Another reason you can't outsource recruiting is that recruiting takes a tremendous amount of creativity. Otherwise, you're going to be doing the same cookie cutter stuff that every other company in the world is doing, and you're going to end up with the same interchangeable talent every other company has. So, don't take the fact that you don't know anything about recruiting as a negative.

完全同意。在我最近的公司里，我认为我招募到了迄今为止合作过的最好的团队，并且打破了很多规则。每一次招聘，我们都必须打破一些核心的招聘规则。我不打算一一列举，因为其中一些技巧在当前环境下仍然有效，有些可能是在挑战边界，但我们打破了所有规则。我们打破了关于通勤的反对意见。我们打破了关于‘哦，我要生孩子了’的反对意见。我们打破了关于‘哦，我负担不起行权’的反对意见。我们打破了关于‘哦，但我还在大学’的反对意见。我们甚至打破了那些可能有诸如‘哦，我想被最好的科学家包围，或者我想在一种不同的环境中工作’这样目标的人的反对意见。在 2025 年的环境下，每个人都在拼命招募 AI 人才和工程师。对顶级工程师的需求比以往任何时候都高，因为他们通过新工具获得了巨大的杠杆效应。从给顶尖人才的薪资待遇中就能看出来。你必须有惊人的创造力。你必须打破规则。这也是你不能外包招聘的另一个原因，因为当你外包时，你是在外包给一个不知道哪些规则可以打破、哪些不能打破的人。他们的人力资源部门或者他们自己害怕打破一些你不希望他们打破的规则。但作为创始人，你可以打破关于股权结构的规则。你可能给每个员工一定数量的股票，但可能会遇到一个你必须为他打破规则的人。或者你必须说服他们为什么不能为他们打破规则。或者你必须以不同的方式构建他们的股票补偿，或者以某种方式安排他们的薪水、入职日期、工作时间、地点、头衔、与谁合作、向谁汇报、办公室是什么样的、他们可以雇佣谁、他们在产品或公司的哪个部分有发言权，或者他们如何跨部门扮演角色。你必须打破规则才能得到最优秀的人，因为最优秀的人不是机器里的齿轮。他们不适合一个整洁舒适的位置。他们是跨学科的，他们必须暂时披上一个身份，比如‘我是电气工程师’、‘我是软件工程师’、‘我是营销人员’、‘我是产品经理’等等，但优秀的人有能力做任何事情，他们选择专注于某件事，但他们的输入在任何地方都有价值，并且受到一群同行的尊重。我的联合创始人喜欢用一个拉丁短语‘primis inter paris’——平等中的第一位。他们所有人都是平等的，但每个人凭借其专业知识和特定技能，被其他人公认为在某个领域是平等中的第一位。所以一个人可能更擅长命名和品牌，另一个人可能更擅长工业设计，还有一个人可能更擅长电气工程或软件工程，或者在另一个领域有品味。但一个伟大的团队会有跨学科的人，每个人都能做很多工作，但专精于特定领域，并具有极端的品味和判断力。他们的同行足够聪明，能够认识到他们在哪里拥有这种品味，并赋予他们在那个领域做决策的能力。你不仅要在招聘上打破规则，还要在公司运营和结构上打破规则。每个好公司都是独特的。它的文化是独一无二的。你不能简单地移植它。举个例子，在我最新的公司，我们不用 Slack，这是一种群聊平台。尤其是在小公司，Slack 变成了一个闲逛的地方。有点像电子邮件。在电子邮件中，为一大群人生成任务太容易了。我可以花 5 秒钟发出一封包含 20 个待办事项的邮件，然后其他人要花一整天来回复它。这创造了一种不对称的浪费他人时间的能力。随着时间的推移，电子邮件已经退化为一个信号很低、噪音很大（即垃圾邮件）的媒介。即使是来自朋友、家人和同事的好意垃圾邮件也是如此。所以，我们也转向了短信，我们知道它的进入门槛更高。比如，如果你要给我发短信，最好是很重要的事情。如果你发了很多短信但不太重要，我可能会屏蔽你。或者如果你在一个群聊里，有人开始发太多消息，你就会退出群聊，这就是为什么所有大型群聊最终都会消亡，因为优秀的人会屏蔽它们并离开。Slack 和群聊平台也有类似的动态，随着时间的推移，它们退化为人们随意提问、预测、搞政治、争吵、谈论与工作无关的事情的组合，基本上变成了群体文化建设的娱乐平台，噪音信号比很高。而如果你没有 Slack，如果你有问题，你必须真正思考并尝试自己解决。如果解决不了，你必须找出公司里谁可能知道答案，然后找到那个人，这非常打扰人，而且你必须弄清楚如何恰当地接近他们。你可以说沟通成本太高了。这限制了你公司扩张的能力，而这正是关键。当一小群聪明的人一起努力把产品从 0 做到 1 时，你最不想要的就是规模。规模是你的敌人。只需要一小群人就能创造出伟大的东西。每个好的创始人都知道这一点。史蒂夫·乔布斯在苹果实施保密的原因之一，就是防止团队之间过度交叉、互相干涉、试图抢占功劳。这也是为什么他把 Macintosh 团队和 Apple 2 团队分开安置在不同的楼里。埃隆·马斯克鼓励人们走出会议室，只开站立会议。杰夫·贝佐斯将团队限制在‘两个披萨团队’的规模。

Absolutely. In my own most recent company, I think I've recruited the best team I've ever worked with by far and have broken so many rules. Every single hire, we had to break some core rule of recruiting. I won't go into all of them because some of those are still tricks that are valid in the environment. Some of those are probably pushing boundaries, but we break every rule. We break all the objections around commuting. We'll break the objections around, 'Oh, I'm having a kid.' We'll break the objections around, 'Oh, I can't afford to exercise these options.' We'll break the objections around, 'Oh, but I'm at a university.' We'll even break the objections around people who may have goals like, 'Oh, I want to be surrounded by the best scientists, or I want to work in a different kind of environment.' In the 2025 environment, everyone is trying so hard to recruit AI people and engineers. The demand for top level engineers is higher than ever because they're so leveraged through the new tools. And you can just see that in the salary offers that are going out to the top people. You just have to be incredibly creative. You just have to break rules. This is another reason why you can't outsource recruiting because when you outsource it, you're outsourcing to someone else who doesn't know what rules they can break and what they can't. Their HR or they're afraid that they'll break some rule that you're not going to be happy with them breaking. But as a founder, you can break rules around the cap table. You might be giving a certain amount of stock to each employee, but one might come along for whom you have to break the rules. Or you have to convince them why you can't break the rules for them. Or you have to structure their stock compensation in a different way or their salary in a certain way or their start date or their hours or their location or their title or who they're working with or who they're reporting to or what their office is like or who they get to hire or what say they get to have in what part of the product or in what part of the company or how they get to straddle roles across different parts of the company. You're going to have to break rules to get the best people because the best people are not cogs in a machine. They don't fit into a neat and comfortable place. They're multidisciplinary and they have to temporarily dawn an identity of oh I'm an electrical engineer I'm a software engineer I'm a marketer I'm a product manager whatever but the great people are capable of anything they chose to specialize in a particular thing but their input is valuable everywhere and they're respected by a group of peers my co-founder has a phrase he likes to use from Latin primis inter paris first among equals where they're all peers but each person given their expertise and their particular knowhow is acknowledged by the others as being the first among equals in a given domain. So one person might be better at naming and branding. One person might be better at industrial design. Other person might be better at electrical engineering or software engineering or have taste in a different domain. But a great team will have multidisciplinary people who are each capable of doing many jobs but are specialized and have extreme taste and judgment in particular areas. and their peers are smart enough to recognize where they have that taste and will confer upon them the ability to make decisions in that area. You have to break the rules not just on recruiting but also an operator of the company and how it's structured. Every good company is idiosyncratic. Its culture is unique to it. You can't just transplant it. To give you an example, at my latest company, we don't use Slack, which is one of these group chat platforms. And especially in a small company, Slack just becomes a hangout spot. It's kind of like email. In email, it's too easy to generate tasks for large groups of other people. I can fire off an email with 20 todos for other people and it takes me 5 seconds to generate that email and then it takes up the day of the other people trying to respond to it. It creates this asymmetric ability to waste other people's time. Over time, email has degenerated into a medium where there's very low signal and a lot of noise aka spam. Even well-meaning spam from friends, family, and co-workers. So, we also switch to text messaging where we understand the barrier to entry is higher. Like if you're going to text me, it better be important. And if you're texting me a lot and it's not that important, I'm probably going to mute you. Or if you're in a group text and someone starts messaging too much, you exit the group text, which is why all these large group texts die out over time cuz the good people mute them and leave. Slack and group messaging platforms have a similar dynamic where over time they degenerate into a combination of people asking random questions into thin air, people prognosticating, people politicking, people bickering, people talking about stuff that is not germane to the work and they become largely entertainment platforms for group culture building, which is fine with a high noise to signal ratio. Whereas if you don't have Slack, if you have a question, you have to really think about and try to solve it yourself. And if you can't, you have to figure out who in the company might have an answer to that question. And you have to track that person down, which is highly interruptive. And you have to figure out how to approach them properly. You could argue that communication overhead is too high. This limits your ability to scale as a company, which is exactly the point. When you have a small number of brilliant people working together to try to take a product from 0 to one, the last thing you want is scale. Scale is your enemy. It just takes a small group of people to create something great. Every good founder knows this. One of the reasons Steve Jobs implemented secrecy at Apple was to prevent teams from cross-pollinating too much and being in each other's business and trying to take credit for each other's work. It's also why he moved the Macintosh team into a separate building from the Apple 2 team. Elon Musk encourages people to walk out of meetings and do standing meetings only. Jeff Bezos limits teams to two pizza teams.

这些都是创始人试图将公司去规模化，分解成更小的组件，这样人们才能真正完成工作，而不是把所有时间花在开会和搞政治上。Slack 打破了这些界限。它允许 50 个人同时在一个群里，互相浪费彼此的时间。如果你强迫人们深思熟虑地进行互动，你就会从会议日程转变为创造者日程，这样人们就可以拥有不间断的自由时间来发挥创造力。而创造力才是最重要的，因为我们确实生活在无限杠杆的时代，AI 和机器人技术每天都在让这一点更加清晰。你需要让你的员工感到无聊，而不是忙碌。总是让他们忙于无意义的工作是无效的。你需要给他们创造者时间、建设者时间，这意味着大量不间断的自由时间，用于进行深度创造性工作。然后，当他们从那种状态中探出头来，感到无聊时，他们可以去跑步，或者花时间陪伴家人，甚至可以去刷 TikTok。没有人会评判，但他们可以更好地管理自己的时间。而 Slack 则把会议的弊病变成了全天候的常态。所以现在，除了检查你的电子邮件收件箱，你还得检查你的 Slack 收件箱，它有着类似 TikTok 那种阴险的上瘾循环——里面有很多垃圾信息，但偶尔会有一个 nugget。所以你不断地在这堆垃圾中翻找，试图找到那个 nugget。人们可以通过发送一条消息来不对称地浪费彼此的时间，而这条消息需要其他 50 个人去筛选，判断它是不是垃圾。所以，尤其是在一个小团队中，一对一的沟通要好得多。通过限制 Slack 的使用，尤其是在早期，你可以迫使团队保持小规模。当团队很小时，那些不称职的人就无法隐藏。作为领导者，你可以更好地管理团队。你可以直接管理他们，或者直接与他们合作。这样你才能真正交付改变世界的产品。

These are all attempts by founders to unscale the company to break it down to smaller components so people can actually get work done instead of spending all their time in meetings and politicking. Slack breaks those boundaries. It allows 50 people to be in a group at once and waste each other's time. If you force people to be thoughtful about their interactions, you move from a meeting schedule to a maker schedule, then people can have uninterrupted free time to be creative. And creativity is all that matters because we do live in the age of infinite leverage and AI and robotics are making that more clear every day. You need to let your people be bored rather than busy. Always keeping them busy with make work is not effective. You need to give them makers time, builder time, which means large amounts of uninterrupted free time to do deep creative work. And then when they stick their heads out of that and they're bored, they can go for a run or they can spend time with their family or they can even go surf TikTok. No one's judging, but they get to manage their time better. Where Slack takes the disease of meetings and makes it pervasive 24/7. So now on top of checking your email inbox, you have to check your Slack inbox and it has that TikTok-like insidious addiction loop where there's a lot of slop in there, but once in a while there's a nugget. So you're constantly now running through this pile of slop to find that nugget. People can asymmetrically waste each other's time by sending one message that then 50 other people have to sift through and figure out if it's flop or not. So, especially in a small team, one-on-one communications are much better. And by limiting the use of Slack, especially early on, you can force the team to stay small. And when a team is small, people who aren't pulling their weight can't hide. You can curate it much better as a leader. You can manage them directly or work with them directly. And you can actually deliver world-changing products.

这让我想起了纳西姆·塔勒布的观点：永远不要雇助理，因为这会给你扩大规模的机会，而助理最终会产生矛盾的效果，让你更忙而不是更轻松。

It reminds me of Nassim Taleb's idea of never hiring an assistant because it gives you the opportunity to expand your scale and the assistant ends up having a paradoxical effect of making you busier instead of less busy.

2025 年 7 月，你发推说，初创公司的工作就是发现未被发现的人才，并将其提炼成产品。显然，产品愿景是存在的。显然，你必须找到人才。关键在于“未被发现”。这是我们没有谈到的部分，我认为很多人都忽略了这一点。如果你能轻易地从远处识别出人才，那么其他人也能。你必须在别人之前找到他们。怎么做呢？埃隆可能是这方面的现代大师。尽管乔布斯、奥特曼和其他一些人也在这方面做得非常出色。埃隆使用的策略很有意思。首先，你选择一个极其大胆的使命。一天只有那么多小时。反正你都要工作，不如做点大事。最优秀的人内心深处知道这一点。他们想做大事。例如，我认为最优秀的人不想做电子游戏或浪费人们生命的垃圾娱乐。他们不想建一个加密赌场。最优秀的人想做有意义的工作，因为他们内心深处意识到自己的潜力。所以当他们看到有机会以工程师、艺术家（希望两者兼具，因为我认为伟大的工程师往往也是伟大的艺术家）的身份表达自己时，他们会被大使命所吸引。所以埃隆做的第一件事就是选择一个宏大的使命，并以尽可能大的方式呈现它。我们不是去太空，不是去月球，我们要去火星。这是一个大使命。同样，山姆·奥特曼坚持：我们不只是为了视频流而构建 Sora，我们不只是构建聊天机器人，我们在构建 AGI。他对此毫不动摇。他想构建 AGI。埃隆不想止步于电动汽车，他甚至不想止步于自动驾驶汽车。他想要机器人，他想要一支 1 亿机器人的大军。特斯拉要一路走下去。所以，这些是鼓舞人心的。这些吸引了最优秀的人。第二，你要早。你提出这些使命，并且比别人更早去做。所以埃隆在太空变得酷之前很久就创办了 SpaceX。人们认为那是不可能的。因此，他成功地在别人之前从 NASA、波音、洛克希德和大学中吸引了最优秀的工程师。现在，如果你在一个更拥挤的领域，你需要有创意，找到那个领域里未被发现的人才。等到某人在 Twitter 上出名时，再招募他们就太晚了。所有人都知道了。甚至等到某人有了 pedigree，赢得了所有奖项和论文，也很难招募。你必须想办法接触到他们。所以，要成为一个伟大的招聘者，你首先必须是一个伟大的巫师。而一个伟大的巫师是未被发现人才的优秀猎手，这意味着你必须有品味，必须对他人感兴趣，并且必须投入时间。例如，我的联合创始人喜欢找那些喜欢捣鼓东西的人。他喜欢找奇怪的项目，不是主流的，不是显而易见的。他不看谁在训练更好的 AI 模型，那太明显了。相反，他可能会看一些边缘的东西，比如谁真的在用奇怪的机器学习算法做微观天气预报。然后他会花一两天时间浏览他们的 GitHub 或论文，真正理解它。然后他会深入思考。之后他会带着一个调整或修改回来，给那个人发邮件说：“嘿，我看了你的代码，看了你做的，我觉得非常有趣。我写了一点代码，我想你可能想整合或插入进去。”或者“我有个问题。”通常那是个好问题，一个经过深思熟虑的问题。而那个人会回应得很好，因为他们正在一边捣鼓，有人注意到了他们的捣鼓，并提出了一个好问题。最棒的是，他这样做不是为了招募人。他这样做只是因为那是他为了好玩而做的事。他真正感兴趣。所以，他找到了这些在边缘捣鼓的奇怪独行侠，然后我再去招募他们。大多数时候会失败，有时会成功，但你会找到非常有趣的人。这是一个例子，说明他的品味如何让他找到特定类别的人。所以，你确实需要在未被发现的地方寻找人才。我们最近在公司雇了一个助理，只是我在餐厅遇到的人，非常热情好客，从未在科技公司工作过一天。但你可以看出这个人做任何事都很出色。他们碰过的每件事都质量上乘、有品位，而且他们很用心。我们招募了他们。我们冒了个险。所以，关键在于发现未被发现的人才，而不是显而易见的人才。

In July 2025, you tweeted that the job of a startup is to find undiscovered talent and distill it into a product. Obviously the product vision is there. Obviously you have to find talent. The key is undiscovered. That's the part that we haven't talked about and that I think a lot of people miss. If you can identify the talent from afar easily, so can everybody else. You have to find them before other people do. How do you do this? Elon is probably the modern master of this. Although Jobs, Altman, and a few other people have also done extremely well in this regard. The playbook that Elon uses is interesting. First, you pick a mission that's extremely audacious. There's only so many hours in a day. You're going to work anyway. You might as well work on something big. The best people know that deep down. They want to work on something big. For example, I think the best people don't want to build video games or slop entertainment that's wasting people's lives. They don't want to build a crypto casino. The best people want to do meaningful work because deep down they're aware of their potential. And so when they see an opportunity to express themselves as engineers, as artists, hopefully as both, because I think the great engineers are often great artists as well, they're going to be drawn to a big mission. So the first thing Elon does across the board is he picks a big mission and he frames it in the largest way possible. We're not going into space. We're not going to the moon. We're going to Mars. That's a big mission. Similarly, Sam Altman stays true to we're not just building Sora to video feeds. We're not just building chat bots. We're building AGI. He's not wavering from that. He wants to build AGI. Elon doesn't want to stop at just electric cars. He doesn't even want to stop at self-driving cars. He wants robots. He wants an army of 100 million robots. Tesla's going all the way. So, these are inspiring. These attract the best people. Second, you're early. You lay out these missions and you do it before everybody else does. So Elon did SpaceX long before space was cool. People thought it was impossible. And so he managed to attract the best engineers out of NASA and Boeing and Lockheed and universities before everybody else did. Now if you're in a more crowded space, you need to get creative and find the undiscovered talent in that space. By the time someone's famous on Twitter, it's too late to recruit them. Everybody knows. Even by the time someone is pedigreed, they've won all the awards and the papers. Very hard to recruit. You have to hack your way to them. So to be a great recruiter, you have to first be a great sorcerer. And a great sorcerer is a good hunter of undiscovered talent, which means you have to have taste and you have to have interest in other people and you have to put in the time. For example, my co-founder loves to find tinkerers. He loves to find weird projects, not mainstream projects, not the obvious stuff. He's not looking at who's training a better AI model. That's too obvious. Instead, he might be looking at something adjacent like who's really into using weird ML algorithms for micro weather forecasting. Then he'll spend a day or two going through their GitHub or going through their paper and really understanding it. And then he'll go off and he'll think deeply about it. And then he'll come back with a tweak or a modification and he'll email that person and say, "Hey, I saw your code. I saw what you've done. I thought it's really interesting. I wrote a little bit of code that I think you may want to incorporate or plug in." Or, "I have a question." And usually it's a good question. It's a considered question. And the person responds well because here they are off tinkering on the side and somebody has spotted their tinkering and has a good question about it. And the best part is he's not doing this to recruit people. He's doing this because that's just what he does for fun. He's genuinely interested. So, he finds these weird loners tinkering at the edge and then I get to go in and recruit them. Most of the time it fails and sometimes it succeeds, but you find really interesting people. That's an example of how his taste allows him to source a particular category of people. So, you do have to look for talent in undiscovered places. We recently hired an assistant at the company and it was just someone I ran into at a restaurant who was incredibly hospitable and had never worked a day in their life at a tech company. But you could just tell this person was good at everything they did. Everything they touched was quality and stylish and they cared. We recruited them. We took a chance. So, it's about finding undiscovered talent, not the obvious talent.

外包招聘的另一个问题是，你把招聘交给猎头或 HR，他们没法给你带来怪才。风险太高，他们自己也没有品味。创造者能识别其他创造者，建造者能识别其他建造者，工程师能识别其他工程师，优秀的销售能识别其他销售，文案能识别其他文案。所以，很难外包这件事。顺便提一句，我特别反感招聘那些没有证据证明自己能营销自己的市场和公关人员。比如，你想招人做社交媒体，他们自己最好有一个很棒的社交媒体账号。他们应该在自己的社交媒体领域玩到顶尖水平。事实上，我认为最优秀的社交媒体人员根本雇不到。你只能在他们还非常原始、账号还很小很年轻、正在崛起时发现他们，或者与他们签约，因为他们知道真正的机会是围绕自己建立独特的频道，他们会短暂地租给你他们的频道，而不是完全交给你。

And this is another problem with outsourcing recruiting, which is you hand to a recruiter, you hand to HR. They can't bring you weird people. It's too high risk. They don't have the taste themselves. Makers have taste in other makers. Builders have taste in other builders. Engineers have taste in other engineers. Good salespeople have taste in other salespeople. Copywriters have taste in other copywriters. So, it's very hard to outsource that. As a related aside, one of my pet peeves is hiring marketing and PR people who have no evidence of being able to market themselves. For example, if you want to hire someone to do your social media, they better have a great social media account for themselves. They should be playing their own social media game at the top of the game. And in fact, I would argue the best social media people are not even hireable. You have to discover them when they're very raw, when their accounts are small and young and up and coming, or you have to contract with them because they know that their real opportunity is to build a channel around themselves uniquely and they'll briefly rent you their channel rather than hand it over to you completely.

Angelus 在人才寻源方面做了一件有创意的事，就是把一楼改成了咖啡馆，叫 Founders Cafe。我们这里有源源不断的创始人，一人公司、两人公司，他们刚刚起步，很多公司不会成功，如果他们的公司失败了，我们就有机会招募他们。这不是任何猎头能想出来的主意。

One thing Angelus has done to be creative in sourcing is to turn our first floor into a cafe. It's called Founders Cafe and we have a constant stream of founders, one-man companies, two-person companies that are just trying to get started, and a lot of these companies are not going to go anywhere and we will have the opportunity to recruit them if their companies fail. This is not an idea that any recruiter is going to come up with.

我再进一步。我觉得你开这样的咖啡馆，不应该是因为你想招人，而只是因为你喜欢和创始人待在一起，喜欢有个咖啡馆。这样会更真实、更真诚，不会感觉像工作，你会做得更好，然后还会有附带的好处。

And I would go one level further. I think you should open a cafe like that, not because you want to recruit people, but just because you like hanging around founders and you like having a cafe. That's going to be a lot more genuine and authentic and won't feel like work and you'll do a better job and then there'll be ancillary benefits to it.

同意。我们开这个咖啡馆是因为我们从事服务创始人的事业。2025 年 8 月，你发推说每个伟大的工程师也是艺术家。我深有体会，但让我们也谈谈什么是艺术。我对艺术的定义比传统定义宽泛得多。我认为艺术是那些为自身而做、做得很好、常常创造美感或强烈情感的东西。很多工程师是内向者。顺便说一句，我讨厌“incel”这个词。它只是贬低内向者的方式，可以说是新时代的“书呆子”。如果有人称某人为 incel，我更可能想面试那个人。所以，我们远离这些侮辱性词汇吧。内向者倾向于通过其他事物表达自己，而不是直接出去表达。那他们会做什么？他们会通过手艺表达自己，创造艺术。在我现在的公司，至少一半的工程师有严肃的业余艺术作品，世界级的作品。从优雅的数学证明到美丽的计算机艺术，到用黏土雕刻、设计服装、设计门把手、水瓶。有一个人做了不可思议的音乐视频。非常棒的东西。我看到很多优秀的工程师摆弄 AI 艺术产品，甚至比所谓的艺术家还多。我认为很多艺术家害怕 AI 艺术产品，说这会取代我。而一个没有艺术家身份、不感到威胁的人，它只是一个工具，他们尝试看看它能创造什么。任何为自身而做、尽可能做好的东西都是艺术。伟大的工程师也是艺术家。他们无所不能，只是他们选择了成为工程师，专注于建造东西。因为工程是把你的想法和艺术变成实际工作、有用、体现知识、可重复、人们能从中获得效用的能力。但这并不意味着它不能是美的。

Agreed. We opened it because we are in the business of serving founders. In August 2025, you tweeted that every great engineer is also an artist. I know this experientially, but let's also talk about what art is. My definition of art is much broader than a conventional definition. I characterize art as something that is done for its own sake and done well and often creates a sense of beauty or some strong emotion. And a lot of engineers are introverts. As an aside, I hate the term incel. It's just a way of putting introverts down. It's the new nerd, if you will. If someone says that somebody is an incel, I'm more likely to want to interview them. So, let's move away from the slurs. Introverts tend to want to express themselves through other things rather than going out and expressing themselves directly. So what are they going to do? They're going to express themselves through their craft. They're going to create art. In my current company, at least half the engineers have serious artwork they've done on the side, world-class artwork. Everything from elegant mathematical proofs to beautiful computer art to literally sculpting things with clay, designing clothing, designing doorknobs, water bottles. There's one who's done incredible music videos. Really good stuff. And I see a lot of the better engineers tinker with the AI art products much more so than even so-called artists do. I think a lot of artists are scared by AI art products saying this is going to replace me. Whereas someone who doesn't have that identity of an artist and doesn't feel threatened by it, it's just a tool and they try it out to see what it can create. Anything done for its own sake and done as well as one possibly can is art. And great engineers are also artists. They're capable of anything. It's just they've chosen to be engineers and focused on building things. Because engineering is the ability to turn your ideas and your art into things that actually work, that do something useful, that embody some knowledge in a way that it can be repeated and people can get utility out of it. But that doesn't mean that it can't be beautiful.

我再次引用我的联合创始人的话，如果你问他最好的艺术形式是什么，绘画、音乐、文学等，对他来说，是工业设计。比如，看看 AirPods，它们被空气动力学地雕刻，同时还要能在机器上按规格在中国以一定价格制造，它们令人满意地咔嗒一声吸进充电盒的磁力凹槽。它们开创了内置“查找我的 AirPods”的整个充电盒。它们隐藏了所有额外按钮。它们能携带备用电池。它们贴合你的耳朵，配有可更换的耳塞。这是艺术和工程的奇迹。需要不可思议的艺术性来设计它，使其贴合雕刻般的人耳——一种美丽自然的形态——同时还能以一定价格大规模生产，并让所有小元素协同工作。当你合上 AirPods 的盖子时，会发出非常令人满意的啪嗒声。周围的曲线是 G3 曲线，是手工雕刻然后扫描的，计算机无法生成。拿在手里的感觉，像一颗从天而降的光滑抛光鹅卵石。它是美的东西，是艺术品。我认为人们本能地知道这一点，这就是为什么他们涌向苹果产品而不是各种安卓产品，因为它们被雕刻得像艺术品。用心投入其中，你能感受到。苹果作为一家由真正深切关心的工程师艺术家组成的公司而成功。这就是为什么直到今天，即使所有其他创始人，甚至那些最近市值可能超过苹果的人，仍然都尊敬苹果。我这一代以及我认为许多后续世代的每一位企业家，都最尊敬史蒂夫·乔布斯和他的团队，因为他们真正是艺术家，而不仅仅是工程师。

Again, I'm channeling my co-founder, but if you ask him what is the best form of art, painting, music, literature, etc. for him, it's industrial design. For example, if you look at the AirPods, the way they're sculpted aerodynamically and still have to be manufacturable on a machine at a certain price point by someone in China, according to a spec, the way they satisfyingly click into the little resting places in their case with magnets. The way they pioneered that whole charging case with the Find My AirPods product built in. The way they hid all the extra buttons. The way they made it carry spare batteries. The way they fit inside your ear with the replaceable tips. That is a marvel of art and engineering. It took incredible artistry to figure out how to design it so it fits sculpted into the human ear, which is a beautiful and natural form, while still being mass-producible at a certain price point and making all of the little elements work together. When you close the lid on the AirPods, it makes a very satisfying snap. The way the curves around it, those are G3 curves. Those are hand sculpted and then scanned in. Computers can't generate those. The way it feels in your hand. It feels like a smooth, polished pebble that fell from the sky. It's a thing of beauty. It's a work of art. And I think people intrinsically know that, which is why they flocked to Apple products over various Android products because they are sculpted like works of art. The care goes in there and you can feel it. Apple triumphed as a company of people who genuinely deeply cared of engineer artists. That's why to this day, even all the other founders, even ones who might have built more market cap recently than Apple has, they still all look up to Apple. Every entrepreneur from my generation and I think many subsequent generations looks up to Steve Jobs and his team more than any other because they were truly artists, not just engineers.

对我来说，任何角色的理想人选是：懂技术、是艺术家、不断产生新知识，最后通过代码、产品或 AI 自动化工作中重复的部分。例外情况存在，但任何角色的理想候选人都应该要么拥有这些能力，要么渴望获得这些能力：懂技术、是艺术家、不断产生新知识（称之为创造力），并自动化工作中乏味的部分。

To me, the ideal person for any role is technical, an artist, constantly generating new knowledge, and finally automating the repetitive parts of their job through code, product, or AI. Exceptions apply, but the ideal candidate for any role should either have these capabilities or be aspiring to gain these capabilities: technical, an artist, constantly generating new knowledge, call that creativity, and automating the tedious parts of their job.

我认为没错。值得注意的是，当你谈到自动化时，你排除了通过流程或人员自动化。那是最糟糕的自动化形式，因为那样会把非技术或非创意人员加入流程，这些人不会长期对工作满意，因为他们只是机器中的齿轮，最终会被某种技术取代。这也会改变环境，因为人类是社会性动物。

I think that's right. And it's telling that when you talked about automation, you left out automating through process or people. That's the worst form of automation because then that adds non-technical or non-creative people into the process and those people aren't going to be happy in those jobs for long because they're cogs in the machine and will eventually be replaced by some piece of technology. It also changes the environment because humans are social animals.

如果你开始把不同的人混在一起，他们总是会去迁就别人，所以对话的层次就会下降。比如，如果一屋子政治家和一屋子工程师在一起，你们不会长时间讨论工程问题。最终，你们会滑向共同话题。而在足够大的群体里，共同话题永远是旅行和食物，因为这些话题没有威胁性。人们总是会退化到那个层面。如果你真的想要一个使命导向的强文化，就不能混合太多不同类型的人。这就是邪教和文化的来源。早期团队看起来确实像邪教。他们偏执狂热，他们古怪，但他们的古怪方式相似。如果你开始混合太多不同类型的人，你只会得到一个平淡的平均值，这不是打造伟大公司或产品的方式。这是一个均值回归问题。实际上，有一个著名的创始人（我不点名）在 Quora 上说过，早期公司最不想要的就是所谓的“多样性”。你想要的是一个单一文化，所有人都相信同样的事情。因为如果没有这一点，你就会把所有时间花在争论上。而早期创业公司没有那个时间。所以你需要每个人都已经完全同步。然后，只有少数几件事值得争论，那些真正能推动业务绩效的事情。创始人和其他人一样想受欢迎。所以对外他们会试图表现出共识驱动的形象，有时甚至愚蠢到自己也信了。但我近距离或远距离观察过的每一位伟大创始人，都极其有主见，几乎是以独裁的方式运营公司。早期团队也有主见，他们构建的产品也有主见。有主见意味着他们对产品应该做什么、不应该做什么有强烈的愿景。如果你没有强烈的愿景，最终就会得到一堆相互竞争的功能，一团糟。Jack Dorsey 有句名言：限制细节数量，让每个细节完美。这在消费产品中尤其重要。你必须极其有主见。消费领域所有最好的产品都是通过简洁实现的。你可以说，ChatGPT 和类似 AI 聊天机器人最近的成功，是因为它们比 Google 更简单。Google 看起来是你能构建的最简单的产品——只是一个搜索框。但即使那个框也有局限性。你被训练得不能像聊天一样跟它说话。你输入关键词，还得小心措辞。你不能直接问一个问题然后得到合理的答案。它不会做同义词匹配，然后给你一堆复杂的结果。你得筛选，分辨哪些是广告、哪些是真实结果、排序是否正确，然后还得点进去阅读。ChatGPT 和聊天机器人进一步简化了这一点。你就像跟人说话一样跟它交流。用语音或打字，它直接给你答案。它可能不总是正确，但足够好，而且以文本、语音或图像等形式直接给你答案。所以，它把互联网上我们曾认为最简单的产品——Google——变得更简单了。你永远无法把产品做得足够简单。要做到简单，你必须极其有主见。你必须移除所有不符合你对产品应该做什么的看法的东西。你必须精心移除每一次点击、每一个多余的按钮、每一个设置。实际上，设置菜单里的东西表明你放弃了对用户的责任。给用户选择就是放弃你的责任。也许出于法律或重要原因，你可以保留少数几个，但你应该努力抵制用户必须做出的每一个选择。在 TikTok 和 ChatGPT 的时代，这一点比以往更明显。人们不想做选择。他们不想有认知负担。他们希望你想出正确的默认值，告诉他们应该做什么、看什么，然后呈现给他们。

If you start mixing them together, they're always going to want to accommodate the other people and so the level of conversation will shift. For example, if you have a bunch of politicians in a room and a bunch of engineers, you're not going to be talking engineering for long. Eventually, you will drift into common topics. And in a large enough group, the common topics are always travel and food because they're non-threatening topics. People always degenerate to that. If you really want to have a strong culture of people who are mission-oriented, you can't mix too many different kinds of people. That's where the cult and culture comes from. Early teams do look like cults. They are monomaniacal. They are weird, but they're all kind of weird in a similar way. And if you start mixing too many different kinds of people, you're just going to get a bland average, which is not how you're going to build a great company or product. It's a regression to the mean problem. There's actually an old Quora thread by a famous founder that I won't name where he says the last thing you want in an early stage company is quote unquote diversity. You want a monoculture of people who all believe the same things. Because if you don't have that, you're going to just spend your time arguing about everything. And you don't have that time at an early stage startup. So you need everybody to already be on the same exact page. And then there's a few things that you might argue about that really move the needle on the performance of the business. Founders want to be popular like everybody else. So externally they'll try to project this image that they're consensus driven and sometimes they're even stupid enough to fall for it. But every great founder I've seen up close or even from afar is highly opinionated and they're almost dictatorial in how they run things. Also early stage teams are opinionated and the products they built are opinionated. Opinionated means they have a strong vision for what it should and should not do. If you don't have a strong vision of what it should and should not do, then you end up with a giant mess of competing features. Jack Dorsey has a great phrase, limit the number of details and make every detail perfect. And that's especially important in consumer products. You have to be extremely opinionated. All the best products in consumer land get there through simplicity. You could argue the recent success of ChatGPT and similar AI chatbots is because they're even simpler than Google. Google looked like the simplest product you could possibly build. It was just a box. But even that box had limitations on what you could do. You were trained not to talk to it conversationally. You would enter keywords and you had to be careful with those keywords. You couldn't just ask a question outright and get a sensible answer. It wouldn't do proper synonym matching and then it would spit you back a whole bunch of results that was complicated. You'd have to sift through and figure out which ones were ads, which ones were real, were they sorted correctly, and then you'd have to click through and read it. ChatGPT and the chatbot simplified that even further. You just talk to it like a human. Use your voice or you type and it gives you back a straight answer. It might not always be right, but it's good enough and it gives you back a straight answer in text or voice or images or whatever you prefer. So, it simplifies what we looked at as a simplest product on the internet, which was formerly Google, and makes it even simpler. And you just cannot make a product that's simple enough. To be simple, you have to be extremely opinionated. You have to remove everything that doesn't match your opinion of what the product should be doing. You have to meticulously remove every single click, every single extra button, every single setting. In fact, things in the settings menu are an indication that you've abdicated your responsibility to the user. Choices for the user are an abdication of your responsibility. Maybe for legal or important reasons, you can have a few of these, but you should struggle and resist against every single choice the user has to make. In the age of TikTok and ChatGPT, that's more obvious than ever. People don't want to make choices. They don't want the cognitive load. They want you to figure out what the right defaults are and what they should be doing and looking at, and they want you to present it to them.

沃伦·巴菲特说你应该雇佣有精力、智慧和正直的人。Joel Spolsky 换了个说法：你想要聪明且能成事的人。在智力方面，对我来说真正重要的只有一个关键测试：候选人是否在创造新知识——这不过是“他们是否有创造力”的另一种说法。否则你只是在雇佣一个机器人，而它的工作本应被自动化。

Warren Buffett says that you should hire people who have energy, intelligence, and integrity. Joel Spolsky put it another way that you want people who are smart and get things done. What's become important to me on the intelligence side is that there's really only one significant test for a candidate, which is are they generating new knowledge, which is just a fancy way of saying are they creative? Because otherwise you're just hiring a robot whose job should be automated.

我认为这很正确。人们可能会不高兴，说你在叫他们机器人。但我不认为有人想一遍又一遍做同样的事。每个人都想做新的、独特的、有创意的事。每个人都可以成为艺术家——不是拿画笔绘画那种，而是创造新知识并享受这个过程。只是领域不同而已。甚至搞明白如何利用 Twitter 或 YouTube 传播你的想法，也是一种创造新知识。比如，几年前传播想法的方式可能是写博客。现在可能是 X 加 Substack，或者在 TikTok 上走红，或者做一个精彩的创业发布视频。目标一直在变，人们总是试图运用创造力去破解那个系统。甚至融资也是如此。与其一个一个去见 VC，我认为如果你的产品展示得好，不如做一个杀手级的发布视频和出色的产品演示，试图在网上走红，展现个性，从噪音中脱颖而出。所以创造力可以应用于任何地方。我寻找的另一个特质是自我驱动。你不需要告诉他们该做什么。你不需要推他们，问“这周你完成了什么？”那是 Elon 著名的提问，我认为很好，但本质上仍然是管理问题。是经理的问题，不是领导者的问题。领导力是激励人。你给他们前进的方向，但一旦他们知道大家共同的方向，他们就会相对自我驱动。他们会找出最佳方式到达那里，并为团队做出贡献。如果他们需要被告知何时前进，需要被推着走、被鞭策，那他们就不适合早期创业公司。

I think that's correct. And people may get unhappy saying, well, you're calling these people robots. But I don't think anybody wants to do the same thing over and over. Everybody wants to do new, unique, and creative things. Everyone can be an artist. Not in the sense of grabbing a paintbrush and painting, but in the sense of creating new knowledge and enjoying that process. It may just be in different domains. Even figuring out how to hack Twitter or YouTube to get your word out is a form of creating new knowledge. For example, a couple of years ago, the way to get the word out was probably writing blog posts. Now, it might be X plus Substack or going viral on TikTok or doing a great startup launch video. The target is always moving and people are always trying to apply creativity to hack that system. Even fundraising. Rather than going and meeting VCs one by one today, I would argue you're better served if your product shows well to build a killer launch video and a great demo of the product and try to get it to go viral online to have a personality and stand out from the noise. So creativity can be applied anywhere. The other thing I look for in people is being self-motivated. So you don't have to tell them what to do. You don't have to push them, hey, what did you get done this week? That's a famous Elon question and I think it's a great one, but it is fundamentally still a management question. It's a manager's question. It's not a leader's question. With leadership, you motivate people. You give them the place to march to, but they're relatively self-motivated once they know what direction you're all headed in. They're going to figure out how best to get there and contribute to the team getting there. And if they have to be told when to march and they have to be pushed along and flogged, then they don't belong in an early stage startup.

所以我认为自我激励非常重要，正如我之前提到的，低自我也很重要，这些是相当罕见的组合，但高自我的人会破坏团队的运作。

So I think being self motivated is really important and as I'd mentioned earlier low ego is also very important and these are pretty rare combinations but high ego people can just destroy the functioning of a team.

很多原则在公司达到一定规模后就很难采用了。例如，你可以设定标准说我只和自驱的人合作，只和不需要太多指导的人合作。这样我就能和尽可能多的人一起工作，而不必时刻盯着他们。但如果公司已经有四五十人，很可能你已经招了一些这样的人。那你要怎么办？大规模裁员？依据什么？一些关于动力的模糊感觉？你的信念有多强？可能不够。

A lot of these principles are very difficult to adopt once the company is past a certain size. For example, you could have the criteria saying I'm only going to work with self-motivated people. I'm only going to work with people who don't need a lot of direction. So, I can work with as many of them as possible and I don't have to look over their shoulder all the time. But if you already have 40 or 50 people in the company, odds are you've already hired a bunch of these people. Then what are you going to do? A mass layoff based on what? Some fuzzy feeling about motivation. How much conviction do you have? Probably not enough.

所以，你建立的团队就是你建立的公司。创始人的个性就是公司，因为你的原则、你的底线和你的价值观决定了你会雇佣谁。

So, it's not only that the team you build is a company you build. Literally the founders's personality is the company because your principles and your non-negotiables and your values dictate who you're going to hire.

最好的创始人对人和产品有极端的品味。他们非常挑剔。例如，在我现在的公司，我对投资者有极端的品味。我不会接受任何风投的钱。我不尊重大多数风投。他们只是到处推钱的资金管理者。很多人喜欢把别人的功劳占为己有。我过去和风投有过不愉快的经历。没有哪个风投能坐在我的董事会里给我建议，那些建议我可能早就听过了。那我为什么要招募风投呢？我对风投会有非常极端的品味。我的联合创始人对建设者有极端的品味。我对营销人员、销售人员和文案有极端的品味。我永远不会雇佣一个不能直截了当的营销人员。这种人很少见。挑剔是有帮助的。你确实需要非常有主见。任何告诉你倾听他人、建立共识和收集反馈的人，都是在暗示你很弱，你做得不够好，或者你的方法不对。好的创始人非常有主见。问题是坏的创始人也非常有主见。

The best founders have extreme taste in people and in products. They are extremely judgy. For example, in my current company, I have extreme taste about investors. I won't take money from any VC. I don't respect most VCs. They're just money managers pushing money around. A lot of them like taking credit for other people's work. I've had bad encounters with VCs in the past. There's no VC who's going to sit on my board and give me advice, which I probably haven't already heard. So, what am I going to recruit a VC for? I'm going to have very extreme taste about a VC. My co-founder has extreme taste about builders. I have extreme taste about marketers and sellers and copywriting. I'm never going to hire a marketing person who can't outright me. And that's a rare person. It helps to be very judgy. You do want to be very opinionated. Anyone who tells you to listen to others and build consensus and gather feedback is implying that you're weak, you're not good enough at what you do, or that you have the wrong approach. Good founders are incredibly opinionated. The problem is the bad founders are very opinionated, too.

人们有很多方法来评估一个人是否有创造新知识的能力。彼得·蒂尔有一个著名的问题：什么重要的真理很少有人同意你？他试图找出那个人是否有自己的观点。他们有自己的想法吗？我有时会问人们，他们对自己的爱好有没有提出过独特的理论。即使是关于壁球，我也听过有人提出关于壁球的独特理论。如果你能产生新知识，你在学习壁球的第一小时内就会开始想出关于如何打和教壁球的想法。它们可能不全对，但你会提出新颖的理论。纳瓦尔最近在推特上也提到了一个问题，我总结为：“你关心什么不流行的事情？”这是另一种评估一个人是否有能力产生自己想法的方法。

There's a lot of ways people try to assess whether someone has the ability to create new knowledge. Peter Teal has his famous question, what important truth do very few people agree with you on? He's trying to find out if that person has opinions of their own. Do they have their own ideas? I will sometimes ask people whether they have any unique theories that they've come up with about their hobbies. Even if it's about squash, I've heard people give me unique theories about squash. If you are able to generate new knowledge, you will start coming up with ideas about how squash should be played and taught within the first hour of learning squash. They might not all be right, but you will come up with novel theories. Naval also has one question he mentioned on Twitter recently, which I would sum up as, "What do you care about that isn't popular?" That's another way of trying to assess whether that person has the ability to generate their own ideas.

蒂尔那个关于秘密的著名问题，其实是站在投资者的角度，他在寻找企业所做的独特赌注，因为他不想有竞争。正如他所说，竞争是留给失败者的。我们在基础微观经济学中学到，竞争会将利润降至零。他想做一个独特的赌注，或者像早期投资者迈克·梅普尔斯喜欢说的那样，“非共识且正确”。但那是针对投资的。我认为在日常工作中，你想和那些善于不断从工作中提炼洞察的人一起工作。

Teal's famous question about a secret is really from an investor's perspective where he's hunting for the unique bet that the business is making because he doesn't want competition. As he says, competition is for losers. As we learn in basic microeconomics, competition reduces profits to zero. And he wants to make a unique bet or as Mike Maples early investor likes to say, non-consensus and right. But that's for investing. I think in everyday work, you want to work with people who are very good at distilling the insight from their work on a constant basis.

马尔科姆·格拉德威尔推广了一万小时定律。一万小时方向上是正确的，但并不完全准确。它暗示如果你花一万小时做某事，你就会精通。我们先不管一万这个数字是否准确。这不仅仅是投入的时间，而是迭代的次数。你有多少个学习循环来推动学习曲线？什么是迭代？迭代就是你做一件事，然后看结果，以某种方式测试结果，理想情况下是针对自由市场、自然或物理。然后你问这个实验的哪部分有效或无效。基于此，你对如何改进做出新的创造性猜测，然后再做一次。你能进行这种循环、迭代的次数越多，你学得就越快。这就是你想要的学习曲线。伟大的人会从每次迭代中提炼洞察。所以这不仅仅是找到一个秘密那么简单。是的，每家公司都有一个秘密赌注。他们有一个关于世界将如何发展的理论，其他人不一定有，或者还不是传统智慧。但在这个过程中，他们会发现成千上万的洞察，每一个都建立在前一个之上，而这一切都由他们能进行的迭代次数驱动。

Malcolm Gladwell popularized 10,000 hours. 10,000 hours is directionally correct, but it's not exactly correct. It implies that if you spend 10,000 hours doing something, you get mastery. Let's put aside whether 10,000 is the right number or not. It's not just hours put in, it's iterations. How many learning loops do you have that drive the learning curve? What is an iteration? Iteration is when you do something and then you look at the result, you test the results somehow, ideally against a free market, nature or physics. Then you ask what part of this experiment worked or not. And then based on that, you make a new creative guess on how to improve that thing and you do it again. The number of times you can do that rotation, that iteration, the faster you're going to learn. That's the curve you want to be on. Great people will distill insights from every iteration. So it's not as simple as finding one secret. Yes, every company makes a secret bet. They have a theory as to how the world is going to work out that other people don't necessarily have end mass or it's not conventional wisdom yet. But along the way, they're going to discover thousands of insights and each one will build upon the last and that's all going to be driven by the number of iterations they can do.

在扩大创业公司规模时遇到的一个问题是，你雇了一个人，这个人又雇了另一个人，而那个人习惯了大公司里定义明确的工作。他们习惯于因工作获得认可。他们与最终结果脱节，以至于他们只需要给经理留下好印象。这是委托代理问题。所以他们现在希望自己的工作不被丢弃。当公司规模超过 20、30、40、50 人时，你经常会听到一个反对意见：“我们不想尝试这个，因为它可能行不通。”这实际上是创始人随着公司规模扩大而面临的头号问题：他们想出的点子比组织能执行的多，而且内部会有阻力，因为十个点子中有九个是不成熟的。但实际上，你处于一个搜索过程、学习过程、发现过程中，你在试图找到有效的方法，你必须尝试很多东西。一个好的创始人会有能力在许多事情上迭代，并丢弃那些无效的，因为学习必然包含失败。记住，所有新信息一开始都是错误信息。它一开始并不明显正确，所以被指责为错误信息。最终随着时间的推移，它被证明是对是错。如果是对的，那么它就是你可以在此基础上构建的信息。一个好的创始人会为此挣扎。我的建议是坚持下去。弄清楚你的组织完成事情的能力。让人们接受这样一个想法：他们的大部分工作都会被丢弃。一切都是实验，丢弃也没关系，只要你在过程中提炼洞察，就要习惯反复的小失败。巴吉·斯里尼瓦桑对此有另一种说法：在想法迷宫中漫步。

One of the problems you run into when scaling a startup is you hire someone who hires someone and that person is used to a well-defined job at a larger company. They're used to getting credit for their work. They're divorced enough from the end outcome that all they have to do is kind of impress their manager. The principal agent problem. So now what they want is for their work to not be thrown away. A common objection you get when your company scales beyond 20, 30, 40, 50 people is, "We don't want to try this because it's probably not going to work." It's actually probably the number one thing a founder will struggle with as a company scales that they'll come up with more ideas in their organization can execute upon and there will be internal resistance to doing things because nine out of 10 ideas are halfbaked. But really, you're in a search process, you're in a learning process, you're in a discovery process, you're trying to find the thing that works and you do have to try a lot of things. And a good founder will have the ability to iterate on many things and throw away the things that didn't work because learning necessarily involves failure. Remember all new information starts as misinformation. It starts as not being obviously true. And so it's accused of being misinformation. Eventually over time it's proven right or wrong. If it's right then it's information you can then build upon. A good founder will struggle with exactly this. My advice would be power through it. Figure out what is your organizational capacity to get things done. Get people comfortable with the idea that most of their work is going to be thrown away. It's all experimentation and it's fine for it to be thrown away and get comfortable with repeated small failure as long as you distill the insights along the way. Baji Shrinavasan has another way of putting this which is wandering through the idea maze.

你左转右转、回溯、摸索什么行得通什么行不通。可能大致方向还是你最初出发的那个方向，但认为你总是朝着正确方向前进，这是一种自我膨胀。这里最大的障碍是骄傲。人们固守最初的愿景，没有好好地在想法迷宫中导航。关键在于反复尝试、回溯、走岔路，直到找到出路。这就是为什么从外部看，一家公司做的事情似乎微不足道，竞争对手很容易赶上——常见的情况是，一家初创公司脱颖而出，然后你说：‘那家大公司会碾压它。’不。只要这家初创公司持续在想法迷宫中探索，它实际上比大公司深入得多。即使大公司模仿它，等大公司到达初创公司所在的位置时，初创公司早已领先，处于迷宫的不同部分。大公司无法抗拒探索那些初创公司多年前就已探索过并知道是死胡同的侧道。这种快速迭代、从中学习、不断产生新洞察和秘密的能力，才是成功的秘诀。这不仅仅是问创始人‘你相信什么别人不相信的事情’那么简单。而是每一天你都会发现新东西，建立在旧东西之上，然后意识到：哦，事情并不像我想象的那样运作，它们实际上是另一种方式。

You're taking left turns and right turns and backtracking and figuring out what works and what doesn't. It might be in the rough direction where you started out. Although it's an ego trip to think that you're always going to be moving in the right direction. The biggest impediment here is pride. People stay locked into their original vision and they don't properly navigate the idea maze. It's about taking lots and lots of repeated steps and backtracks and side turns until you find your way through the maze. This is why even though from the outside it looks like what a company has done is trivial and it's going to be easy for competitors to catch up to them and it's a common thing to see a startup break out and then you say, 'Well, that big company's just going to crush it.' No. As long as that startup keeps wandering through the idea maze, they're actually much deeper down through the maze than the big company is. Even if the big company copies them, by the time it gets to where the startup is, the startup has moved way ahead. It's in a different part of the maze. The big company can't resist the urge to explore side hallways that the startup has already explored years ago and knows are dead ends. It's this ability to iterate very quickly and to learn from it and constantly generate new insights and secrets. That is a secret to success. It's not just the one simple secret where you ask the founder, what is the thing you believe that nobody else does. It's literally every single day you figure out something new that builds upon something old and you realize, oh, things don't work like I thought they would. They actually work a different way.

你必须愿意容忍的一个副作用是，优秀的团队会抛弃大部分工作。在我看来，大多数人招聘中缺失的要素是‘不容忍’。你真的应该把公司里的每个员工，包括你自己，都视为试图通过引入平庸人才来摧毁公司的敌方间谍。不幸的是，这只是人性。

The side effect that you have to be willing to tolerate is that great teams are throwing away most of their work. To me, the missing ingredient in most people's recruiting, is intolerance. You should really just treat every employee in the company, including yourself, as an enemy agent that's trying to destroy the company by bringing mediocre talent into the business. It's unfortunately just human nature.

我和联合创始人在公司里定了一个新标准：只要天才。这个词很苛刻，但它设定了很高的门槛。现在你只需环顾四周，看谁不是天才。吸引天才的唯一方法，无论这个词对你意味着什么，就是让公司里全是天才。如果有人不是天才，那么要么你正进入一个不再能雇佣天才的阶段，出于某种原因需要规模化；要么你需要让那个人走人，因为你过早地雇佣了他们，不符合你想建立的公司类型。这非常困难。一个月能招到一个天才就算幸运了。作为创始人，你必须识别他们，并尽一切努力招募和激励他们。所以这本质上是自我限制的，因为一个人可能不会在公司待超过三、四、五年——尽管在一些伟大的公司，人们会待几十年。按这个流失率，你只能维持 30 到 50 人的公司。但如果你能组建一个 10 人的天才团队，你就遥遥领先了。在大多数成功的公司里，创始人和少数早期员工是天才级别的。但在急于规模化的冲动中，这一点很快就被淹没了。我认为‘天才’这个词甚至有点被低估了。我认为每个人都有自己擅长的领域。你要找的是那些已经找到自己擅长领域的人，或者有能力、有潜力在为你工作期间找到或接近那个领域的人。作为投资者，我也有一个不公平的优势。我经常和那些在公司里没成功、我不得不让他们走的人共事，但我足够了解他们，能识别他们的擅长领域，然后说：‘这不是你发挥所长的地方，但如果你以后做别的事情，告诉我，我可能会投资。’这在几个案例中效果不错。所以你说得对，人们往往只需要合适的环境。你无法改变的是动力。如果一个人没有动力，不想全力以赴，或者生活中有其他事情，那你就只能在这个节点上放弃他们。一个较少被提及的事情是，你常常会在错误的时间遇到正确的人。他们有内部问题、生活问题、家庭问题、健康问题，这些事情使他们无法达到你需要的水平。这很遗憾，但经常发生。相关地，人们说：‘哦，我精疲力尽了，需要休息一两个月充电。’根据我的经验，这基本不成立。通常，精疲力尽是你正在做某件要么不成功、要么你根本不喜欢的工作的信号。单纯休息解决不了问题。如果你真的喜欢你所做的事，它通常会给你更多能量和动力。少数情况，比如我知道埃隆·马斯克以鞭策团队到凌晨 4 点、在深夜开员工会议、进行疯狂的死循环而闻名。那是他设定和建立的文化。这没问题。在这种情况下，我能看到某些人精疲力尽。但即使在那里，他们说的是：我未来无法承受这种工作量。所以即使在那里，休息也没用，因为当你回来时，他会像之前一样给你任务。所以一般来说，当有人说‘我精疲力尽了’，我就理解为‘我想辞职’，即使他们自己未必意识到。

My co-founder and I have a new criterion in our company, geniuses only. It's a harsh word, but it sets a very high bar. Now you can just look around for who's not a genius. The only way you're going to attract geniuses, whatever that term means to you, is by having a company full of geniuses. And if someone's not a genius, then either you're transitioning into the phase where you can no longer hire geniuses and you just need to scale up for whatever reason, or you need to show that person the door because you hired them prematurely for the kind of company you're trying to build. Now, this is very difficult. You're lucky if you can hire one genius a month. You as a founder have to identify them and do whatever it takes to recruit them and motivate them. So it's inherently self-limiting given that a person probably isn't going to stick around your company for more than 3, 4, 5 years. Although in some great companies, people stick around for decades. At that attrition rate, you're talking about a 30 to 50 person company. But if you can even assemble a team of 10 geniuses, you're way ahead of everybody else. At most companies, the successful ones, the founders and maybe a few early people are at the genius level. But in the urge and the rush to scale, that gets drowned out too quickly. I think genius is actually even a bit of an underused term. I think everybody does have a zone of genius. You want to find people who have already found their zone of genius or they have the capability, they have the slope to be able to find their zone of genius or get close to it while they're still working at your company. As an investor also, I have an unfair advantage. I've often worked with people where it hasn't worked out in a company and I have to let them go, but I've gotten to know them well enough that I recognize their zone of genius and I can say, 'This is not where you're operating in your zone of genius, but if you ever end up doing this other thing, let me know cuz I'll probably want to invest.' And that has actually worked out reasonably well in a couple of cases. So, you're right. People often just need to be in the right environment. The thing you can't fix is motivation. If someone's just unmotivated, if they don't want to apply themselves fully, if they have other things going on in their life, then you just have to cut them off at this point. One of the things that's less talked about is often you'll meet the right person at the wrong time. They just have internal problems, life problems, home problems, health problems, things that are going on that make them not capable of functioning at the level that you need. And that's a sad situation, but it happens all the time. On a related note, people say, 'Oh, I'm burned out. I need to take a break for a month or two and recharge.' In my experience, that's largely not true. Usually, burnout is a sign you're working on something that either isn't working or you don't enjoy the work fundamentally. Just taking time off won't fix it. If you're really enjoying what you do, generally that'll give you more energy and more motivation. There are rare cases like I know Elon is famous for flogging his teams until 4 in the morning and calling staff meetings at odds of the night and doing crazy death marches. That's the culture that he sets and builds. That's fine. In those situations, I could see certain people burning out. But even there, what they're saying is, I cannot sustain this workload in the future. So even there, taking time off doesn't work because when you come back, he's going to put you to task the same way as earlier. So generally, when someone says, 'I'm burned out,' I just read that as, 'I want to quit.' even if they don't necessarily realize it themselves.

你必须小心你带进组织的人，因为他们会不知不觉地带来自己的审美观。他们会不知不觉地雇佣像他们自己的人。例如，在 AngelList，团队中有个人在决定雇佣两个顾问时，选了错的那个，因为他觉得：‘我和这个顾问在一起会更开心。’那个有趣的人在个性和举止、沟通方式上更像他们。我心目中的‘有趣’是开发优秀产品并取得成功。大卫·多伊奇会这样说：‘当你觉得有趣时，你正在能力边缘学习。’如果你不觉得有趣，那意味着什么？你没有学到新东西。你没有在学习。如果它引起焦虑，那意味着什么？那意味着它超出了你的能力。所以，如果你在能力边缘运作，你就处于心流状态。你在学习、在做事，压力足够大以至于有趣，但又不至于大到焦虑。

You have to be careful about who you bring into the organization because they will bring their own sense of aesthetics without even knowing it. They will hire people that are like them without knowing it. For example, at AngelList, one of the people on the team was trying to decide between two consultants that we wanted to hire, and he was picking the wrong one because he was like, I think I'll have more fun with this consultant. The fun one was more like them in their personality and their way of carrying themselves and communicating. My idea of fun is working on great products and succeeding. David Deutsch would say something like, 'When you're having fun, you're learning at the edge of your capability to learn.' If you are not having fun, what does that mean? You're not getting anything new. You're not learning. If it's anxiety-inducing, what does that mean? That means it's beyond your capability. So, if you're operating at the edge of your capability, you're in flow. You're learning, you're doing, you're being stressed enough for it to be interesting, but not so stressed that you're anxious.

而且这很有趣。可能每时每刻并不有趣，但当你回顾每一天、每一周、每一个月时，它是很有趣的。还有什么比在你能力的最高水平上磨练技艺更值得做的呢？所以，我认为“有趣”这个标准也适用于商业和工作。例如，在我最近的公司里，设计师们对设计痴迷到这种程度：我们正在装修新的办公空间，这根本不是他们的工作，也没人要求他们设计，我们甚至可能不希望他们插手，但他们就是忍不住要精心设计到每一个细节。同样，我让他们做一本书来收集我们工作过程中的各个里程碑，结果他们自己设计书脊，自己买打印机用特殊纸张印刷。他们非常执着，绝不会只做到 80%。沃伦·巴菲特 famously 拒绝在高尔夫球赛上下注，因为他不赌博，不冒险，只做有把握的事，这就是他的模式。同样，一个好的工程师不会允许自己写出糟糕的代码。我知道你想务实，想走捷径，想把事情尽快完成，但真正伟大的工程师不会制造粗制滥造的东西。伟大的设计师不会半吊子地设计。我会删除有 1 万点赞的推文，因为我发现了一个语法或拼写错误，或者想到了更好的表达方式。我会直接删掉，不在乎浏览量，因为我希望它完美。有时人们在推特上嘲笑我，因为我发了一条推文，然后改变主意，在它获得大量关注后删除，调换两个词的顺序或改一个词。第二天早上过了编辑窗口，我又觉得原来的更好，就重新发出来，失去了所有病毒式传播和势头。但我不在乎。我不想与草率的言论联系在一起。它必须正确且无可挑剔，必须以有趣的方式表达真实的东西。艺术正确比受欢迎更重要。

And it's fun. It may not be fun moment to moment, but when you look back day to day, week to week, month to month, it is fun. What else would you rather be doing than practicing your craft at the highest level of capability at your edge? So, I do think the fun criterion applies to business and to jobs. For example, at my most recent company, the designers are obsessed with design to the point where we are getting a new office space. It's not their job to design it. Nobody asked them to design it. We probably didn't even want them to design it. They can't help but design it down to a tea. It's meticulous. Similarly, I asked them for a book in which we could just collect various checkpoints along the way of the work that we're doing. The people that we have, they're designing their own book binding. They got their own printer to print it on a special paper. They're obsessive. They can't 80% design something. Warren Buffett famously refused to put a bet on a golf game because he doesn't bet. He doesn't take risks. He only does short things. That's his whole model. The same way a good engineer will not let themselves write a shoddy piece of code. And I know you want to be practical and you want to cut corners and you want to get things out the door. But a truly great engineer is not going to create something shoddy. A great designer is not going to halfway design something. I will delete tweets that have 10,000 likes on them because I catch a grammar or spelling error or I think of a better way to formulate it. I'll just kill it. I don't care about the views because I want it to be done just right. People make fun of me on Twitter sometimes because I'll put out a tweet, then I'll change my mind and I'll delete it after it's gotten a lot of traction and I'll reverse the order of two words or I'll change one word. Then I'll wake up the next morning pass the edit window and I decide I like the original one and I'll post the original one back up and I've lost all the virality and all the momentum. But I don't care. I don't want to be associated with a slipshot statement. It has to be correct and incompressible. It has to say something true to me in an interesting way. And that's more important that the art is correct than that it's popular.

是的。有句老话说，不擅长工作的人，你让他们做某事，他们会尝试去做，然后你还得检查他们的工作。而最优秀的人，你让他们做某事，他们会带回你从未想到、也从未想象过的东西。他们是高能动性的人，创始人模式，随便你怎么称呼，但他们是那种以最佳方式完成工作为己任的人。你只需要向他们传达你认为需要完成的事情。这不仅仅是沟通，沟通是管理层面的事，更是领导力。所以你还得激励他们，不是那种老套的加油打气，而是帮助他们理解你为什么认为这件事如此重要。如果你觉得它真的重要，那么你的工作就是要么同样说服他们，要么被他们说服，因为你自己可能错了。一旦他们相信了重要性，他们能动性足够高，就会以绝对最佳的方式去执行。至于你提到的创造力，他们会在解决问题过程中产生新知识和新创意，并以你意想不到的方式解决。有时你和某人交谈，不真诚的人会抓住你的字眼攻击你，而聪明的人会理解你真正想表达的意图。高智商的人往往会回答你真正想问的问题，而不是你问的那个问题。如果要总结这次对话，创业公司的首要原则就是永远不要在人才上妥协。我宁愿在客户体验上短期受挫，也不愿在团队质量上短期或长期受损。我会用两个词概括：精选人才。我未来在现在公司以及后续公司的理念是，我只想和天才一起工作。我只想和自驱的人、低自我的人、建设者、工程师、艺术家以及各自领域顶尖的人一起工作。就这么简单。你只需要愿意精选人才。顺便说一句，我们还没谈解雇。但那是另一面。你总会犯错。寻找人才很难，招聘很难，领导也很难。我不喜欢“管理”这个词，因为优秀的人不需要被管理。但解雇、让人离开也很难，但你必须做。你永远不会有 100%的成功率，甚至接近都难。如果你不解雇人，那意味着你在稀释自己。所以，你必须让不匹配的人离开。否则，你只会招到比他们更弱的人，公司会慢慢恶化。关于招聘天才的另一个附注：只招天才。这是目前的座右铭。显然这是理想化的，但你不是在填坑，不是在填补职位。这是一个常见的陷阱：“我需要一个营销岗位，所以我要面试一堆营销人员，然后从中选最好的。”不。如果他们不是天才，就不要招。作为创始人，你只需要清楚大致需要哪些能力，然后寻找能填补这些能力的天才。如果你发现一个天才，即使他不具备这些能力，但 somehow 可以雇佣，那就立刻招进来。所以，收集天才，囤积他们。你永远不会后悔。挑战可能在于如何让他们保持兴趣，因为你可能没有合适的岗位。但优秀的人总能找到问题并参与其中，即使那不是他们的“本职工作”。所以当你发现真正优秀的人时，只要可能就招进来，不管有没有空缺。试图把优秀的人塞进钉孔、方块、三角或圆洞里是错误的。真正的天才非常独特，彼此不像，无法被装进盒子。通过填补职位来招聘，本质上就是试图把人塞进盒子。所以，我甚至认为你不一定需要为职位招聘。是的，你希望人们拥有对公司重要的技能，但优秀的人比我们人为制造的僵化盒子灵活得多，那些盒子更多是 HR 和大公司的产物。

Yeah. There's an old quote that people who are not good at their jobs, you ask them to do something, they try and do it and then you have to check their work. The best people, you ask them to do something and they come back with something that you never could have come up with yourself and never could have imagined. It's high agency people, founder mode, whatever you want to call it, but it's people who take responsibility for doing the job the best way possible. You just have to communicate to them what it is that you think needs to be done. And it's not just communication. Communication is a management thing. It's a leadership thing. So you also have to motivate them, not in some cheesy rahrra way, but to help them understand the insight you have as to why you think it's so important. And if you think it's really important, then it's your job to either convince them equally that it's important or to be talked out of it yourself because you might have made a mistake. And then once they're convinced it's important, they're high agency enough that they will just go and do it in the absolute best way possible. And to your point about creativity, they'll come up with new knowledge and new creativity along the way to figure out how to solve the problem. And they'll solve it in a way you didn't even know. Sometimes you're in a conversation with someone and a disingenuous person is going to latch on to the exact words you said and jump on you. Whereas a smart person is going to understand the intention of what you're actually trying to say. And a highly intelligent person will often answer the question not that you asked, but the question you really wanted to ask or you meant to ask. If I was going to sum up this whole conversation, the prime directive of a startup is to never compromise on talent. I would rather take a shortterm hit on customer experience than take a short or long-term hit on the quality of the team. I would summarize it in two words, curate people. And the philosophy that I have going forward in my current company and all subsequent ones is that I only want to work with geniuses. I only want to work with self-motivated people. I only want to work with low ego people. I only want to work with people who are builders and engineers and artists and are at the top of their craft. And that's all there is to it. You just have to be willing to curate people. We haven't talked about firing, by the way. But that's the other side of it. You will always make mistakes. Sourcing is hard. Recruiting is hard. Leadership is hard. I don't like the word management because great people don't need to be managed. But firing, letting go of people is hard, too. But you have to do it. You're never going to have a 100% hit rate. Not even close to it. If you're not firing, it means that you're diluting yourself. So, you do need to let people go who don't match up. Otherwise, you're only going to recruit people who are weaker than them, and your company will slowly deteriorate. One other side note on hiring geniuses. Only hire geniuses. That's the current motto. Obviously, aspirational, but you're not trying to fill slots. You're not trying to fill roles. That is a common trap you fall into. Well, I need to fill a marketing role, so I'm just going to interview a bunch of marketing people and then I'll hire the best one out of that set. Nope. If they're not a genius, don't hire them. Just be aware as a founder of what are the rough capabilities you need and then look for geniuses who can fill those capabilities. And if you find a genius who doesn't fill any of those capabilities, but is somehow hirable, hire them right away. So, collect geniuses, warehouse them. You'll never regret it. Your challenge may be to keep them interested because you may not have the right fit for them. But great people have a way of identifying whatever the problem is and getting involved even if it's not their quote unquote job. So when you find someone who's truly great, you just hire them anyway if you can, regardless of whether you have a slot or not. It's a mistake to try and fit great people into pegs and squares and triangles and holes. The real geniuses are incredibly idiosyncratic. They don't resemble each other. You cannot fit them into a box. By trying to fill a role, you're inherently trying to fit somebody into a box. So, I don't even think you necessarily want to hire for roles. Yes, you want people to have skill sets that matter for your company, but good people are much more flexible than these artificial rigid boxes that we make out, which are more of a function of HR and large companies.

小公司不应该采用来自数十亿或数千人规模的大公司做法，比如人力资源、岗位和薪酬等级之类的东西。作为创始人，你总是在破解系统。所以，你必须时刻保持高度灵活。当你发现天才时，就直接招募。

Small companies should not be applying large company practices that come from multi-billion or multi-thousand person companies. Things like HR and roles and compensation brackets and things of that nature. As a founder, you're always hacking the system. So, you always have to be incredibly flexible on your feet. And when you recognize genius, just recruit.