返回列表
🧠 阿头学 · 💬 讨论题

如何更具雄心：给拥有模型的人类的一份教程

本文主张通过提升抽象层级、强制外部验证与跨域视角重构AI工作流，但将人类心理学直接套用于模型交互存在逻辑越界，且严重低估了工程基建与合规成本。
打开原文 ↗

2026-10-15 原文链接 ↗
阅读简报
双语对照
完整翻译
原文
讨论归档

核心观点

  • 高度决定价值上限：从执行单点任务转向对最终业务结果负责是AI产生指数级价值的唯一路径，必须通过开放系统权限与重构数据管道使模型接管目标闭环。
  • 深度依赖原始数据与外部验证：顶尖专家级输出必须基于未加工的原始材料与专业工具链，且必须引入独立于模型的外部评判机制以彻底阻断幻觉自证循环。
  • 广度源于动词重构与多智能体对抗：将交付物名词转化为动作与目的能精准触发跨领域视角引入，但必须以具体动作指令替代空洞的角色扮演，否则必然退化为术语拼盘。
  • 自我效能感是突破使用瓶颈的心理杠杆：设定具体且困难的目标能显著拉升模型输出质量，但人类需主动校准预期以避免陷入“能力通胀下的使用降级”。

跟我们的关联

  • 对 ATou 意味着必须放弃“填空式”提问，下一步应直接套用“名词转动词”框架重写核心业务Prompt，并强制接入自动化验证脚本以替代人工复核。
  • 对 Neta 意味着提示词工程已触及能力天花板，下一步需将MCP连接器与沙箱环境深度集成，以支撑“高度”与“深度”所需的系统级权限调用与工具链编排。
  • 对 Uota 意味着AI不能仅用于降本增效，下一步应利用“广度”策略引入红队视角重构数据看板，将季度复盘升级为实时风险预警与资源动态调配机制。

讨论引子

  • 当模型能力边界受限于架构与训练数据时，“设定不合理目标”的策略在何种阈值下会从“激发潜力”滑向“系统性制造幻觉”？
  • 企业级AI落地中，数据权限与安全审计的刚性约束，是否从根本上否定了“赋予模型跨系统权限以实现高度雄心”的可行性？
  • 在缺乏领域专家介入的情况下，如何量化评估“动词重构”与“多智能体对抗”产出的真实业务价值，而非陷入自嗨式输出？

如何更具雄心：给拥有模型的人类的一份教程

Felix Rieseberg 关于作品文章在线 自动
◂ 文章如何更具雄心：给拥有模型的人类的一份教程

▶︎ 播放0:00:00

↑ ↓ 滚动 · ← → 章节

排版

2026年10月13日 8分钟

如何更具雄心：给拥有模型的人类的一份教程

发布于 2026年10月13日

阅读时间 8分钟

主题

我在 Claude 团队工作。这是一段奇妙的经历：每天我都能了解到借助 Claude 或由 Claude 完成的又一壮举，这些是我未曾设想、认为不可能，或是单纯缺乏好奇心与时间去尝试的。这个月，它或许协助发现了一颗新行星，我看到了第一部我真正喜欢的 AI 音乐视频，而我们正在探讨数学的新纪元。

模型的能力并非均匀增长。正如 Terence Tao 所言，模型“在某些方向上令人惊叹，而在其他方面则显得不足”。有些人能够驾驭这些令人惊叹的能力，仿佛正在翱翔——看着人们利用我参与构建的产品达到不可思议的高度，实在令人着迷。而许多人仍在努力让这对新翅膀运转起来，疑惑除了问“法国的首都是哪里”或“能让这封邮件的语气别那么生气吗”之外，它还能有什么用。

我经常（与其他 AI 从业者一起）宣称，我们在使用它时应该更具雄心。这已成为一句战斗口号。每当看到令人印象深刻的成果，我们就会转发并写道“我们的雄心还不够”。我理解遵循这一建议有多么棘手，因此这里有一份关于如何更具雄心的教程。

我将雄心分为四种类型：速度（Speed）、高度（Altitude）、深度（Depth）与广度（Range）。

  • 速度（Speed）：用更少的时间完成相同的工作，这是 AI 经济价值的基石（bread and butter）：如果你需要写代码，我用 Claude 会更快。这一点已非常成熟，在使用 AI 的方式上几乎不需要创造力，因此我将跳过此项。

  • 高度（Altitude）：提升抽象层级（abstraction），关注最终结果而非单个任务。

  • 深度（Depth）：以领域顶尖专家的视角来审视任务。我们不再停留在任务表层，而是深入一层，触及底层的数据、工具与问题。

  • 广度（Range）：借鉴多个领域的知识，而非仅停留在任务表面所需的那一个领域，运用你平时不会使用的知识、工具与技术。

我主要使用 Claude（显而易见），但我的建议并不局限于特定模型。

高度（Altitude）

我们常轻率地称之为“让模型在更高的抽象层级上运行”。当你给模型分配任务时，它只是在执行你对于如何达成预期目标的最佳猜测。若要在高度上更具雄心，请问问自己——接下来你希望发生什么？目标之上的目标是什么？去年年初，我还会让 Claude 为我修复单个崩溃问题。最终，我直接要求 Claude 确保我们的应用不再崩溃。

让我们考虑一个日常（敢说是平淡无奇？）的办公任务。你在虚构的 Acme Inc 公司，需要为季度业务回顾（QBR）准备一份演示文稿。只要给一堆数据，Claude 会很乐意为你制作幻灯片。也许你甚至有模板，它只需填空即可。这显然无法与发现新行星相提并论，所以让我们看看一些具体步骤，让我们的方法更具雄心。我们将从基础开始，每一步都再提升一点雄心：

  • 作为输入，大多数回顾都想知道本季度发生了什么变化、原因是什么，以及我们该怎么做。克制住只更新模板中现有数字的冲动——尽可能把你能拿到的所有数据都交给模型，让它理解你的业务：“告诉我本季度发生了什么变化、为什么，以及我们该怎么做。”

  • 会议（至少是好的那种）理想情况下应以某些决策告终。请问 Claude：“根据你掌握的所有信息，找出最紧迫的三项决策，并开始针对潜在方向展开正反论证。”

  • QBR 诞生于那个时代：好的回顾需要人类仔细筛选数据，弄清数据含义，并判断业务是否需要调整。请问 Claude：“回顾的存在是为了让领导层尽早发现问题，并将资金和人员调配到正确的位置。设计一种比季度演示文稿更好的方法，用我们自身的历史数据证明它能更早发现问题，并与现有流程并行运行一个季度。”

如果你真正追求的是某个最终结果，高度策略效果最佳；对于创意类工作则效果稍逊。判断是否适合采用高度雄心的一个好迹象是：你是否需要开会讨论你正在做的事情。目前仍存在局限（在 2026 年，“Claude，去赚钱”这句话本身对大多数企业似乎并不奏效），但我普遍观察到，随着模型能力越强，它们似乎越能在更高的抽象层级上运行。

高度也是人们找到持久工作流、将人类移出循环（human out of the loop）的地方，从而减少你不断用一个个任务更新模型的必要性。要做好这一点，通常需要改变你的工作流——为了让 Claude 确保我们的应用不再崩溃，我首先必须确保它能轻松获取崩溃报告，理解哪个应用版本对应哪个 git 提交，并赋予它访问 Windows、macOS 和 Linux 计算机的权限，以便它能真正独立复现并验证修复方案。

深度（Depth）

有时，你只是希望任务被出色地执行。在深度上具备雄心，意味着推动模型使用该领域顶尖专业人士所使用的工具、技术、媒介与知识——并持续推动，直到连那些专业人士都可能从你的工作中有所收获。

让我们继续以 Acme Inc 的季度业务回顾演示文稿为例。假设 Acme Inc 正面临用户流失（随时间推移的客户流失）的困扰。我发现结合以下技巧非常有效：

  • 坚持使用行业顶尖人士所用的工具：当被提示扮演专业人士时，模型往往只切换词汇而不切换方法。要深入解决我们的问题，我们需要使用与顶尖数据科学家和统计学家相同的工具。当 Claude 倾向于使用 Markdown 表格时，坚持要求它使用数据库和统计模型。你不必非得知道专家用什么，Claude 通常知道。

  • 设定看似不合理的目标，并告诉模型它们非常合理：你需要网站在 5 秒而非 10 秒内加载？那就要求 200 毫秒。模型觉得某事不可能？重申这完全可解，它只需继续推进。

  • 挑战你的既有认知：尤其在商业领域，我们深信不疑的许多事物只是对现实的粗略近似。让模型访问原始信息和数据，并要求它对抗性地检验假设。“我们认为流失是新手引导问题。试着从原始数据中复现这一点。”

  • 产生新知识：这很容易附加在挑战认知之后——一旦 Claude 已深入数据内部，就要求它寻找盲点。“构建一个客户模拟模型，根据我们的日志复现过去一年的情况。然后找出我们尚未察觉的最早流失信号，并证明它在未见过的数据上也能做出预测。”

  • 引入外部评判：不要让模型自我评判。我认为随着模型智能的提升，这最终会得到解决，但在 2026 年 10 月，我恳请你坚持要求外部验证。理想情况下，那是一个模型可以运行以立即获得反馈的工具。

一个经常帮到我的简单提示词：

我想要这个任务的更深版本，而不是更大版本。在重写之前：（1）说明该领域的顶尖专业人士在此会做哪些我未要求的事；（2）指出他们会从何种最原始的材料入手，而非已经为我消化过的内容；（3）列出他们会使用的具体方法与工具；（4）说明除了你或我之外，还有什么能证明结果是正确的。然后重写我的提示词，精确要求上述内容。

广度（Range）

简而言之，在广度上具备雄心意味着我们从不同领域的视角来审视工作——这些领域的数量在过去是经济上无法承受的。这并非什么独特的洞见，每一部劫案电影的前 30 分钟都在组建一支角色与才能各异的团队。随着 AI 产品让运行多智能体系统（multi-agent systems）（通过多线程、编排器或简单的子智能体）变得越来越容易，我通过恳求 Claude 引入额外领域，看到了非常有趣的结果。

将其付诸实践并不容易；它需要你的创造力。直到我看到同事 Alex Albert 使用 Blender 渲染室内设计，我才从未考虑过为我的主页使用 Claude 构建 3D 环境——尽管我之前做过电子游戏！然而，这恰恰是我最常看到魔法发生的地方。

那么，我们该如何做到这一点？我使用以下技巧：

  • 尝试用动词替换名词：这是 Tom Kelley 和 Jonathan Littman 在《创新的艺术》（The Art of Innovation）中的一个老技巧。季度回顾不是“一份演示文稿”。它是“让忙碌的人在 30 分钟内相信一个关于因果的故事，并在不确定性下做出决策”。拥有动词后，我和 Claude 都更容易理解哪些领域可能带来独特且有用的视角。

  • 组建团队：这份演示文稿必须惊艳，否则外星人就要炸毁地球了。人类会组建什么样的团队？为了保证制作质量，我会引入分析师、记者和编辑。如果需要演示，可能还需要动态图形与音频设计师。引入对抗性的“红队”（red team）也可能很有用，所以让我们请来做空者（short seller）和法务会计师。

很容易不小心陷入角色扮演，这只会让对话记录变得奇怪，却不会改善实际结果。使用此方法来寻找领域与技术。不要要求模型“扮演一个做空者”，而是要求它去做做空者为研究其头寸所做的事情。

我所描述的雄心实现方式，都要求模型拥有一台能够安装软件并访问各类相关数据的计算机（无论是通过 MCP 连接器（MCP connectors）还是其他方式）。有许多工具可以实现这一点，我建议你使用其中之一。

我们曾身处此地

以上就是我关于如何更具雄心的具体建议。如果你愿意再停留片刻，我想告诉你为什么我认为这首先是有必要的。

我们人类始终被自己认为“可能”的事物所限制——更糟的是，我们通常只在非常狭窄的学科内信任自己。众所周知，Michelangelo 最初并不想绘制 Sistine Chapel 的壁画——你看，他是个雕塑家，不是画家。

在研究雄心问题时，我了解到心理学对此有一个名称：自我效能感（self-efficacy）。简而言之，自我效能感高的人更相信自己的能力，进而实际上能完成更多事情。有证据表明自我效能感可以被“破解”：Locke 和 Latham 仔细研究了 35 年的目标设定研究，自信地宣称，被赋予具体困难目标的人，其表现优于那些仅仅被告知“尽力而为”的人。Stripe 创始人 Patrick Collison 维护着一份“快速”清单，记录了人类在同时获得大胆目标与紧迫期限时所完成的诸多成就。一些有据可查的“破解”方法包括：

  • 通过相信其可能性来完成困难之事：1939 年，George Dantzig 上课迟到，在 Berkeley 的统计学课上抄下黑板上的两道题，并当作作业交上了自己的解答。那是统计学中两道尚未解决的难题。

  • 在目睹他人做到后完成困难之事：几十年来，人类一直在追逐四分钟一英里的目标。Roger Bannister 在 1954 年 5 月打破了它，但在短短 46 天内他仍是唯一做到的人。如果你关注跑步，我们正看到类似的事情发生在两小时马拉松上：今年 4 月，Sabastian Sawe 在伦敦跑出了首个官方认可的两小时内马拉松。在同一场比赛中，亚军也突破了两小时大关。

  • 通过允许自己失败来完成困难之事：显然，每个人最喜欢的商业书籍是《网球的内心游戏》（The Inner Game of Tennis），我上周才刚发现。该书的核心论点是：如果你真的非常非常努力，你就打不出最好的网球。我们工程师可能还记得 Linux，它最初被宣布为“只是一个爱好，不会像 GNU 那样庞大和专业”。

自我效能感并非新问题，但在当下它变得更为强大。在一个我们的能力每隔几周就实现飞跃的世界里，我们需要一些技巧，让我们能够去做那些我们未曾真正考虑或认为可能的事情。如果你发现了更多能让你更具雄心的技巧，请告诉我！

本文配图由 Francisco Goya 创作，《Modo de volar》（一种飞翔的方式）

— 结束 —

Felix Rieseberg 我是 Anthropic 的工程经理（Engineering Manager, EM），迄今为止负责 Claude.ai、Cowork 和 Claude Code Desktop 的工程工作。我是 Electron 及其他多个开源项目的共同维护者。我曾在 Notion、Stripe、Slack 和 Microsoft 工作过。

更多文章 上一篇 由机械手段制造 所有文章

索引下一篇上一篇首页

How to be more ambitious: A tutorial for humans with models

如何更具雄心：给拥有模型的人类的一份教程

Felix Rieseberg AboutWorkPostsOnline AUTO
◂ PostsHow to be more ambitious: A tutorial for humans with models

Felix Rieseberg 关于作品文章在线 自动
◂ 文章如何更具雄心：给拥有模型的人类的一份教程

▶︎ PLAY0:00:00

▶︎ 播放0:00:00

↑ ↓ SCROLL · ← → SECTIONS

↑ ↓ 滚动 · ← → 章节

TYPESET

排版

13 October 20268 MIN

2026年10月13日 8分钟

How to be more ambitious: A tutorial for humans with models

如何更具雄心：给拥有模型的人类的一份教程

Published13 October 2026

发布于 2026年10月13日

Reading time8 min

阅读时间 8分钟

Topics

主题

I work on Claude. It's an interesting experience: Every day I learn about yet another feat accomplished with or by Claude that I hadn't considered, didn't think possible, or simply lacked the curiosity and time to try. This month, it may have helped discover a new planet, I've seen the first AI music video that I really like, and we are discussing a new era of math.

我在 Claude 团队工作。这是一段奇妙的经历：每天我都能了解到借助 Claude 或由 Claude 完成的又一壮举，这些是我未曾设想、认为不可能，或是单纯缺乏好奇心与时间去尝试的。这个月，它或许协助发现了一颗新行星，我看到了第一部我真正喜欢的 AI 音乐视频，而我们正在探讨数学的新纪元。

Model capabilities don't grow uniformly. Models are, as Terence Tao puts it, "astounding in some directions, while inadequate in others". Some people manage to pilot the astounding capabilities and seem to be flying - it's addicting to watch people reach incredible heights with the products I've built. Many others are still struggling to get their new wings to work, wondering what they're good for beyond asking "What is the capital of France" or "Can you make this email sound less angry".

模型的能力并非均匀增长。正如 Terence Tao 所言，模型“在某些方向上令人惊叹，而在其他方面则显得不足”。有些人能够驾驭这些令人惊叹的能力，仿佛正在翱翔——看着人们利用我参与构建的产品达到不可思议的高度，实在令人着迷。而许多人仍在努力让这对新翅膀运转起来，疑惑除了问“法国的首都是哪里”或“能让这封邮件的语气别那么生气吗”之外，它还能有什么用。

I frequently proclaim (together with other people working on AI) that we ought to be more ambitious when using it. It's become a rallying cry. Whenever we see something impressive, we repost and write "we're not ambitious enough". I understand how tricky it is to follow that advice, so here's a tutorial on being more ambitious.

我经常（与其他 AI 从业者一起）宣称，我们在使用它时应该更具雄心。这已成为一句战斗口号。每当看到令人印象深刻的成果，我们就会转发并写道“我们的雄心还不够”。我理解遵循这一建议有多么棘手，因此这里有一份关于如何更具雄心的教程。

I think of ambition in four types: Speed, Altitude, Depth, and Range.

我将雄心分为四种类型：速度（Speed）、高度（Altitude）、深度（Depth）与广度（Range）。

  • Speed: The same work in less time, the bread and butter for AI's economic value: If you need some code written, I'll do it faster with Claude. Well established and requires little creativity in how you approach AI, so I'll skip this one.
  • 速度（Speed）：用更少的时间完成相同的工作，这是 AI 经济价值的基石（bread and butter）：如果你需要写代码，我用 Claude 会更快。这一点已非常成熟，在使用 AI 的方式上几乎不需要创造力，因此我将跳过此项。
  • Altitude: Going up in abstraction and considering the outcome rather than the individual task.
  • 高度（Altitude）：提升抽象层级（abstraction），关注最终结果而非单个任务。
  • Depth: Considering a task the way a field-leading expert would. Rather than operating at the task's layer, we go one level deeper and touch the underlying data, tools, and problems.
  • 深度（Depth）：以领域顶尖专家的视角来审视任务。我们不再停留在任务表层，而是深入一层，触及底层的数据、工具与问题。
  • Range: Drawing on multiple fields rather than just the one the task might need at the surface, using knowledge, tools, and techniques you normally wouldn't.
  • 广度（Range）：借鉴多个领域的知识，而非仅停留在任务表面所需的那一个领域，运用你平时不会使用的知识、工具与技术。

I primarily use Claude (duh) but my advice is not model-specific.

我主要使用 Claude（显而易见），但我的建议并不局限于特定模型。

Altitude

高度（Altitude）

We often flippantly call this "letting the model operate at a higher abstraction". When you give the model a task, it's executing your best guess at how to get the outcome you want. To be more ambitious in altitude, ask yourself - what do you want to happen next? What's the goal beyond the goal? At the beginning of last year, I still asked Claude to fix individual crashes for me. Eventually I asked Claude to simply keep our apps crash-free.

我们常轻率地称之为“让模型在更高的抽象层级上运行”。当你给模型分配任务时，它只是在执行你对于如何达成预期目标的最佳猜测。若要在高度上更具雄心，请问问自己——接下来你希望发生什么？目标之上的目标是什么？去年年初，我还会让 Claude 为我修复单个崩溃问题。最终，我直接要求 Claude 确保我们的应用不再崩溃。

Let's consider an everyday (dare I say mundane?) office task. You're at Acme Inc, an imagined company, where you need to put together a deck for the quarterly business review (QBR). Given a bunch of data, Claude will happily make you some slides. Maybe you even have a template and it just needs to fill it in. That's hardly comparable to finding a new planet, so let's look at some concrete steps to make our approach more ambitious. We'll start basic and get a little bit more ambitious with each step:

让我们考虑一个日常（敢说是平淡无奇？）的办公任务。你在虚构的 Acme Inc 公司，需要为季度业务回顾（QBR）准备一份演示文稿。只要给一堆数据，Claude 会很乐意为你制作幻灯片。也许你甚至有模板，它只需填空即可。这显然无法与发现新行星相提并论，所以让我们看看一些具体步骤，让我们的方法更具雄心。我们将从基础开始，每一步都再提升一点雄心：

  • As input, most reviews want to know what changed this quarter, why, and what should be done about it. Resist the urge to just update whatever numbers are in the template - give the model too as much data as you can get your hands on so it can understand your business: "Tell me what changed this quarter, why, and what we should do about it."
  • 作为输入，大多数回顾都想知道本季度发生了什么变化、原因是什么，以及我们该怎么做。克制住只更新模板中现有数字的冲动——尽可能把你能拿到的所有数据都交给模型，让它理解你的业务：“告诉我本季度发生了什么变化、为什么，以及我们该怎么做。”
  • Meetings (the good kind, anyway) ideally end with some decisions. Ask Claude: "Work out which three decisions are most urgent in the face of everything you've learned, and start working on potential directions with for/against arguments."
  • 会议（至少是好的那种）理想情况下应以某些决策告终。请问 Claude：“根据你掌握的所有信息，找出最紧迫的三项决策，并开始针对潜在方向展开正反论证。”
  • QBRs are from a time when good reviews required humans to carefully sift through data, figure out what the data means, and if the business should adjust. Ask Claude: "The review exists so leadership catches problems early and moves money and people to the right places. Design a better way to do that than a quarterly deck, show from our own history that it would have caught things sooner, and run it beside the current one for a quarter."
  • QBR 诞生于那个时代：好的回顾需要人类仔细筛选数据，弄清数据含义，并判断业务是否需要调整。请问 Claude：“回顾的存在是为了让领导层尽早发现问题，并将资金和人员调配到正确的位置。设计一种比季度演示文稿更好的方法，用我们自身的历史数据证明它能更早发现问题，并与现有流程并行运行一个季度。”

The altitude approach works best if you really are after some ultimate outcome; it works less well for creative endeavors. A good tell-tale sign that ambition in altitude is a good idea is if you're going to have meetings to discuss whatever you're working on. There are currently limits ("Claude, go make money" doesn't seem to work for most businesses by itself in 2026), but I generally see that models seem to be able to operate at higher levels of abstraction the more capable they become.

如果你真正追求的是某个最终结果，高度策略效果最佳；对于创意类工作则效果稍逊。判断是否适合采用高度雄心的一个好迹象是：你是否需要开会讨论你正在做的事情。目前仍存在局限（在 2026 年，“Claude，去赚钱”这句话本身对大多数企业似乎并不奏效），但我普遍观察到，随着模型能力越强，它们似乎越能在更高的抽象层级上运行。

Altitude is also where people find durable workflows that take the human out of the loop, making it less necessary for you to constantly update the model with task after task. Doing this well often involves changing your workflow - in order to let Claude make sure our app is crash-free, I first had to make sure it can easily fetch crash reports, understand which app version maps to what git commit, and give it access to Windows, macOS, and Linux computers so it can actually independently reproduce and verify fixes.

高度也是人们找到持久工作流、将人类移出循环（human out of the loop）的地方，从而减少你不断用一个个任务更新模型的必要性。要做好这一点，通常需要改变你的工作流——为了让 Claude 确保我们的应用不再崩溃，我首先必须确保它能轻松获取崩溃报告，理解哪个应用版本对应哪个 git 提交，并赋予它访问 Windows、macOS 和 Linux 计算机的权限，以便它能真正独立复现并验证修复方案。

Depth

深度（Depth）

Sometimes, you just want a task executed really well. Being ambitious in depth means pushing the model to use the tools, techniques, media, and knowledge employed by the best professionals in the field - and to keep pushing until even those people might learn something new from your work.

有时，你只是希望任务被出色地执行。在深度上具备雄心，意味着推动模型使用该领域顶尖专业人士所使用的工具、技术、媒介与知识——并持续推动，直到连那些专业人士都可能从你的工作中有所收获。

Let's continue with our Acme Inc example of a quarterly business review deck. Assume for a second that Acme Inc struggles with some user churn (customer loss over time). I've found a mix of the following techniques useful:

让我们继续以 Acme Inc 的季度业务回顾演示文稿为例。假设 Acme Inc 正面临用户流失（随时间推移的客户流失）的困扰。我发现结合以下技巧非常有效：

  • Insist on the same instruments used by the best in the industry: When prompted to act like a professional, models often switch vocabulary without actually switching methods. To work on our problem with depth, we need to use the same tools as the best data scientists and statisticians. When Claude reaches for markdown tables, insist on the use of databases and statistical models. You don't necessarily have to know what the experts use, Claude usually does.
  • 坚持使用行业顶尖人士所用的工具：当被提示扮演专业人士时，模型往往只切换词汇而不切换方法。要深入解决我们的问题，我们需要使用与顶尖数据科学家和统计学家相同的工具。当 Claude 倾向于使用 Markdown 表格时，坚持要求它使用数据库和统计模型。你不必非得知道专家用什么，Claude 通常知道。
  • Pick unreasonable goals and tell the model that they're very reasonable: You need your website to load in 5s instead of 10s? Ask for 200ms. Does something seem impossible to the model? Reiterate that it's very much solvable, it just needs to keep going.
  • 设定看似不合理的目标，并告诉模型它们非常合理：你需要网站在 5 秒而非 10 秒内加载？那就要求 200 毫秒。模型觉得某事不可能？重申这完全可解，它只需继续推进。
  • Challenge what you know: Especially in business, a lot of things we believe to be true are rough approximations of reality. Give the model access to the raw information and data and ask it to adversarially check assumptions. "We believe churn is an onboarding problem. Try to reproduce that from the raw data."
  • 挑战你的既有认知：尤其在商业领域，我们深信不疑的许多事物只是对现实的粗略近似。让模型访问原始信息和数据，并要求它对抗性地检验假设。“我们认为流失是新手引导问题。试着从原始数据中复现这一点。”
  • Produce new knowledge: Easily tacked onto challenging what you know - once Claude is already deep inside the data, ask it to find blind spots. "Build a simulation of our customers that reproduces the last year from our logs. Then find the earliest signal of churn we don't know about yet, and show it predicts on data it has not seen."
  • 产生新知识：这很容易附加在挑战认知之后——一旦 Claude 已深入数据内部，就要求它寻找盲点。“构建一个客户模拟模型，根据我们的日志复现过去一年的情况。然后找出我们尚未察觉的最早流失信号，并证明它在未见过的数据上也能做出预测。”
  • Use an outside judge: Do not ask the model to judge itself. I assume that this will eventually be solved with increasing model intelligence, but in October 2026, I implore you to insist on outside validation. Ideally that's a tool the model can run to get feedback immediately.
  • 引入外部评判：不要让模型自我评判。我认为随着模型智能的提升，这最终会得到解决，但在 2026 年 10 月，我恳请你坚持要求外部验证。理想情况下，那是一个模型可以运行以立即获得反馈的工具。

A simple prompt that often helps me:

一个经常帮到我的简单提示词：

I want a deeper version of this task, not a bigger one. Before rewriting it: (1) say what a top professional in this field would do here that I'm not asking for; (2) name the rawest material they would start from instead of what has already been digested for me; (3) name the specific methods and tools they would use; (4) say what, other than you or me, would show the result is right. Then rewrite my prompt to ask for exactly that.

我想要这个任务的更深版本，而不是更大版本。在重写之前：（1）说明该领域的顶尖专业人士在此会做哪些我未要求的事；（2）指出他们会从何种最原始的材料入手，而非已经为我消化过的内容；（3）列出他们会使用的具体方法与工具；（4）说明除了你或我之外，还有什么能证明结果是正确的。然后重写我的提示词，精确要求上述内容。

Range

广度（Range）

In short, being ambitious in range means that we consider our work from the perspective of different fields - more fields than would have been economical in the past. This isn't a particularly unique insight, every single heist movie spends the first 30 minutes assembling a diverse cast of characters and talents. As AI products make it easier and easier to run multi-agent systems (with multiple threads, orchestrators, or simply subagents), I've seen really interesting results by imploring Claude to bring in additional fields.

简而言之，在广度上具备雄心意味着我们从不同领域的视角来审视工作——这些领域的数量在过去是经济上无法承受的。这并非什么独特的洞见，每一部劫案电影的前 30 分钟都在组建一支角色与才能各异的团队。随着 AI 产品让运行多智能体系统（multi-agent systems）（通过多线程、编排器或简单的子智能体）变得越来越容易，我通过恳求 Claude 引入额外领域，看到了非常有趣的结果。

Putting this into practice isn't easy; it requires you to be creative. Until I saw my colleague Alex Albert use Blender to render interior designs, I never considered using Claude for 3D environments for my homepage - even though I've previously made video games! However, this is easily where I see the most magic happen.

将其付诸实践并不容易；它需要你的创造力。直到我看到同事 Alex Albert 使用 Blender 渲染室内设计，我才从未考虑过为我的主页使用 Claude 构建 3D 环境——尽管我之前做过电子游戏！然而，这恰恰是我最常看到魔法发生的地方。

So, how do we do this? I use the following techniques:

那么，我们该如何做到这一点？我使用以下技巧：

  • Try to replace the nouns with verbs: An old trick from The Art of Innovation by Tom Kelley and JonathanLittman. A quarterly review isn't "a deck". It's "getting busy people to believe a story about cause and effect, and to decide under uncertainty, in 30 minutes". Having the verbs makes it easier for both me and Claude to understand which fields might bring in a unique and useful perspective.
  • 尝试用动词替换名词：这是 Tom Kelley 和 Jonathan Littman 在《创新的艺术》（The Art of Innovation）中的一个老技巧。季度回顾不是“一份演示文稿”。它是“让忙碌的人在 30 分钟内相信一个关于因果的故事，并在不确定性下做出决策”。拥有动词后，我和 Claude 都更容易理解哪些领域可能带来独特且有用的视角。
  • Assemble the crew: The deck needs to be amazing, otherwise aliens will blow up Earth. What kind of team will humanity get together? For production quality, I'd bring in an analyst, a journalist, and an editor. If it has to be presented, probably a motion and audio designer, too. It might be useful to employ an adversarial "red team", too, so let's bring in a short seller and a forensic accountant.
  • 组建团队：这份演示文稿必须惊艳，否则外星人就要炸毁地球了。人类会组建什么样的团队？为了保证制作质量，我会引入分析师、记者和编辑。如果需要演示，可能还需要动态图形与音频设计师。引入对抗性的“红队”（red team）也可能很有用，所以让我们请来做空者（short seller）和法务会计师。

It is easy to accidentally engage in role play, which will make your transcript weirder without improving the actual outcome. Use this method to find fields and techniques. Don't ask the model to write a short seller, ask it to do what a short seller does to research their position.

很容易不小心陷入角色扮演，这只会让对话记录变得奇怪，却不会改善实际结果。使用此方法来寻找领域与技术。不要要求模型“扮演一个做空者”，而是要求它去做做空者为研究其头寸所做的事情。

All the ways in which I describe ambition require the model to have a computer with the ability to install software and access to all kinds of relevant data (either using MCP connectors or other means). Plenty of tools make this possible and I recommend that you use one.

我所描述的雄心实现方式，都要求模型拥有一台能够安装软件并访问各类相关数据的计算机（无论是通过 MCP 连接器（MCP connectors）还是其他方式）。有许多工具可以实现这一点，我建议你使用其中之一。

We have been here before

我们曾身处此地

That's my concrete advice for how to be more ambitious. If you're willing to stick around for a minute, I'd like to tell you why I think this is even necessary in the first place.

以上就是我关于如何更具雄心的具体建议。如果你愿意再停留片刻，我想告诉你为什么我认为这首先是有必要的。

We humans have always been held back by what we believe possible - and what's worse, we usually trust ourselves only in very narrow disciplines. Michelangelo famously didn't want to paint the Sistine Chapel - you see, he was a sculptor, not a painter.

我们人类始终被自己认为“可能”的事物所限制——更糟的是，我们通常只在非常狭窄的学科内信任自己。众所周知，Michelangelo 最初并不想绘制 Sistine Chapel 的壁画——你看，他是个雕塑家，不是画家。

While researching the problem of ambition, I learned that psychology has a name for this: self-efficacy. In short, people with high self-efficacy believe in their own abilities more, and in turn actually accomplish more. There's evidence that self-efficacy can be hacked: Locke and Latham pored over 35 years of goal-setting research to confidently proclaim that humans who are given specific hard goals perform better than those who are simply told to try their best. Stripe founder Patrick Collison maintains a "Fast" list, chronicling the many things humans have accomplished when given both audacious goals and tight deadlines. Some hacks that are well documented:

在研究雄心问题时，我了解到心理学对此有一个名称：自我效能感（self-efficacy）。简而言之，自我效能感高的人更相信自己的能力，进而实际上能完成更多事情。有证据表明自我效能感可以被“破解”：Locke 和 Latham 仔细研究了 35 年的目标设定研究，自信地宣称，被赋予具体困难目标的人，其表现优于那些仅仅被告知“尽力而为”的人。Stripe 创始人 Patrick Collison 维护着一份“快速”清单，记录了人类在同时获得大胆目标与紧迫期限时所完成的诸多成就。一些有据可查的“破解”方法包括：

  • Accomplish hard things by believing them to be possible: In 1939, George Dantzig showed up late to a statistics class at Berkeley, copied two problems off the blackboard, and handed in his solutions thinking they were homework. They were two unsolved problems in statistics.
  • 通过相信其可能性来完成困难之事：1939 年，George Dantzig 上课迟到，在 Berkeley 的统计学课上抄下黑板上的两道题，并当作作业交上了自己的解答。那是统计学中两道尚未解决的难题。
  • Accomplish hard things after watching somebody else do them: For decades, humans chased the four-minute mile. Roger Bannister broke it in May 1954 but remained the only human to do so for a mere 46 days. If you're into running, we're watching something similar happen to the two-hour marathon: Sabastian Sawe ran the first official sub-two-hour marathon in London this April. In the same race, the runner-up also broke the two-hour mark.
  • 在目睹他人做到后完成困难之事：几十年来，人类一直在追逐四分钟一英里的目标。Roger Bannister 在 1954 年 5 月打破了它，但在短短 46 天内他仍是唯一做到的人。如果你关注跑步，我们正看到类似的事情发生在两小时马拉松上：今年 4 月，Sabastian Sawe 在伦敦跑出了首个官方认可的两小时内马拉松。在同一场比赛中，亚军也突破了两小时大关。
  • Accomplish hard things by giving yourself permission to fail: Apparently everyone's favorite business book is The Inner Game of Tennis, which I discovered just last week. The book's core thesis is that you're not playing your best game of tennis if you're trying really, really hard. We engineers might remember Linux, announced as "just a hobby, won't be big and professional like gnu".
  • 通过允许自己失败来完成困难之事：显然，每个人最喜欢的商业书籍是《网球的内心游戏》（The Inner Game of Tennis），我上周才刚发现。该书的核心论点是：如果你真的非常非常努力，你就打不出最好的网球。我们工程师可能还记得 Linux，它最初被宣布为“只是一个爱好，不会像 GNU 那样庞大和专业”。

Self-efficacy isn't a new problem, but it is a more powerful one now. In a world where our abilities make leaps every couple of weeks, we need techniques that allow us to do things we didn't really consider or believe possible. If you find more techniques that allow you to be more ambitious, please let me know!

自我效能感并非新问题，但在当下它变得更为强大。在一个我们的能力每隔几周就实现飞跃的世界里，我们需要一些技巧，让我们能够去做那些我们未曾真正考虑或认为可能的事情。如果你发现了更多能让你更具雄心的技巧，请告诉我！

The image for this post is by Francisco Goya, “Modo de volar” (A Way of Flying)

本文配图由 Francisco Goya 创作，《Modo de volar》（一种飞翔的方式）

— END —

— 结束 —

Felix Rieseberg Im an EM at Anthropic, where Ive so far lead engineering for Claude.ai, Cowork, and Claude Code Desktop. Im co-maintainer of Electron and various other open source projects. Ive previously worked on and for Notion, Stripe, Slack, and Microsoft.

Felix Rieseberg 我是 Anthropic 的工程经理（Engineering Manager, EM），迄今为止负责 Claude.ai、Cowork 和 Claude Code Desktop 的工程工作。我是 Electron 及其他多个开源项目的共同维护者。我曾在 Notion、Stripe、Slack 和 Microsoft 工作过。

MORE POSTS Older Made by mechanical means All posts

更多文章 上一篇 由机械手段制造 所有文章

INDEXNEWEROLDERHOME

索引下一篇上一篇首页

How to be more ambitious: A tutorial for humans with models

Felix Rieseberg AboutWorkPostsOnline AUTO
◂ PostsHow to be more ambitious: A tutorial for humans with models

▶︎ PLAY0:00:00

↑ ↓ SCROLL · ← → SECTIONS

TYPESET

13 October 20268 MIN

How to be more ambitious: A tutorial for humans with models

Published13 October 2026

Reading time8 min

Topics

I work on Claude. It's an interesting experience: Every day I learn about yet another feat accomplished with or by Claude that I hadn't considered, didn't think possible, or simply lacked the curiosity and time to try. This month, it may have helped discover a new planet, I've seen the first AI music video that I really like, and we are discussing a new era of math.

Model capabilities don't grow uniformly. Models are, as Terence Tao puts it, "astounding in some directions, while inadequate in others". Some people manage to pilot the astounding capabilities and seem to be flying - it's addicting to watch people reach incredible heights with the products I've built. Many others are still struggling to get their new wings to work, wondering what they're good for beyond asking "What is the capital of France" or "Can you make this email sound less angry".

I frequently proclaim (together with other people working on AI) that we ought to be more ambitious when using it. It's become a rallying cry. Whenever we see something impressive, we repost and write "we're not ambitious enough". I understand how tricky it is to follow that advice, so here's a tutorial on being more ambitious.

I think of ambition in four types: Speed, Altitude, Depth, and Range.

  • Speed: The same work in less time, the bread and butter for AI's economic value: If you need some code written, I'll do it faster with Claude. Well established and requires little creativity in how you approach AI, so I'll skip this one.

  • Altitude: Going up in abstraction and considering the outcome rather than the individual task.

  • Depth: Considering a task the way a field-leading expert would. Rather than operating at the task's layer, we go one level deeper and touch the underlying data, tools, and problems.

  • Range: Drawing on multiple fields rather than just the one the task might need at the surface, using knowledge, tools, and techniques you normally wouldn't.

I primarily use Claude (duh) but my advice is not model-specific.

Altitude

We often flippantly call this "letting the model operate at a higher abstraction". When you give the model a task, it's executing your best guess at how to get the outcome you want. To be more ambitious in altitude, ask yourself - what do you want to happen next? What's the goal beyond the goal? At the beginning of last year, I still asked Claude to fix individual crashes for me. Eventually I asked Claude to simply keep our apps crash-free.

Let's consider an everyday (dare I say mundane?) office task. You're at Acme Inc, an imagined company, where you need to put together a deck for the quarterly business review (QBR). Given a bunch of data, Claude will happily make you some slides. Maybe you even have a template and it just needs to fill it in. That's hardly comparable to finding a new planet, so let's look at some concrete steps to make our approach more ambitious. We'll start basic and get a little bit more ambitious with each step:

  • As input, most reviews want to know what changed this quarter, why, and what should be done about it. Resist the urge to just update whatever numbers are in the template - give the model too as much data as you can get your hands on so it can understand your business: "Tell me what changed this quarter, why, and what we should do about it."

  • Meetings (the good kind, anyway) ideally end with some decisions. Ask Claude: "Work out which three decisions are most urgent in the face of everything you've learned, and start working on potential directions with for/against arguments."

  • QBRs are from a time when good reviews required humans to carefully sift through data, figure out what the data means, and if the business should adjust. Ask Claude: "The review exists so leadership catches problems early and moves money and people to the right places. Design a better way to do that than a quarterly deck, show from our own history that it would have caught things sooner, and run it beside the current one for a quarter."

The altitude approach works best if you really are after some ultimate outcome; it works less well for creative endeavors. A good tell-tale sign that ambition in altitude is a good idea is if you're going to have meetings to discuss whatever you're working on. There are currently limits ("Claude, go make money" doesn't seem to work for most businesses by itself in 2026), but I generally see that models seem to be able to operate at higher levels of abstraction the more capable they become.

Altitude is also where people find durable workflows that take the human out of the loop, making it less necessary for you to constantly update the model with task after task. Doing this well often involves changing your workflow - in order to let Claude make sure our app is crash-free, I first had to make sure it can easily fetch crash reports, understand which app version maps to what git commit, and give it access to Windows, macOS, and Linux computers so it can actually independently reproduce and verify fixes.

Depth

Sometimes, you just want a task executed really well. Being ambitious in depth means pushing the model to use the tools, techniques, media, and knowledge employed by the best professionals in the field - and to keep pushing until even those people might learn something new from your work.

Let's continue with our Acme Inc example of a quarterly business review deck. Assume for a second that Acme Inc struggles with some user churn (customer loss over time). I've found a mix of the following techniques useful:

  • Insist on the same instruments used by the best in the industry: When prompted to act like a professional, models often switch vocabulary without actually switching methods. To work on our problem with depth, we need to use the same tools as the best data scientists and statisticians. When Claude reaches for markdown tables, insist on the use of databases and statistical models. You don't necessarily have to know what the experts use, Claude usually does.

  • Pick unreasonable goals and tell the model that they're very reasonable: You need your website to load in 5s instead of 10s? Ask for 200ms. Does something seem impossible to the model? Reiterate that it's very much solvable, it just needs to keep going.

  • Challenge what you know: Especially in business, a lot of things we believe to be true are rough approximations of reality. Give the model access to the raw information and data and ask it to adversarially check assumptions. "We believe churn is an onboarding problem. Try to reproduce that from the raw data."

  • Produce new knowledge: Easily tacked onto challenging what you know - once Claude is already deep inside the data, ask it to find blind spots. "Build a simulation of our customers that reproduces the last year from our logs. Then find the earliest signal of churn we don't know about yet, and show it predicts on data it has not seen."

  • Use an outside judge: Do not ask the model to judge itself. I assume that this will eventually be solved with increasing model intelligence, but in October 2026, I implore you to insist on outside validation. Ideally that's a tool the model can run to get feedback immediately.

A simple prompt that often helps me:

I want a deeper version of this task, not a bigger one. Before rewriting it: (1) say what a top professional in this field would do here that I'm not asking for; (2) name the rawest material they would start from instead of what has already been digested for me; (3) name the specific methods and tools they would use; (4) say what, other than you or me, would show the result is right. Then rewrite my prompt to ask for exactly that.

Range

In short, being ambitious in range means that we consider our work from the perspective of different fields - more fields than would have been economical in the past. This isn't a particularly unique insight, every single heist movie spends the first 30 minutes assembling a diverse cast of characters and talents. As AI products make it easier and easier to run multi-agent systems (with multiple threads, orchestrators, or simply subagents), I've seen really interesting results by imploring Claude to bring in additional fields.

Putting this into practice isn't easy; it requires you to be creative. Until I saw my colleague Alex Albert use Blender to render interior designs, I never considered using Claude for 3D environments for my homepage - even though I've previously made video games! However, this is easily where I see the most magic happen.

So, how do we do this? I use the following techniques:

  • Try to replace the nouns with verbs: An old trick from The Art of Innovation by Tom Kelley and JonathanLittman. A quarterly review isn't "a deck". It's "getting busy people to believe a story about cause and effect, and to decide under uncertainty, in 30 minutes". Having the verbs makes it easier for both me and Claude to understand which fields might bring in a unique and useful perspective.

  • Assemble the crew: The deck needs to be amazing, otherwise aliens will blow up Earth. What kind of team will humanity get together? For production quality, I'd bring in an analyst, a journalist, and an editor. If it has to be presented, probably a motion and audio designer, too. It might be useful to employ an adversarial "red team", too, so let's bring in a short seller and a forensic accountant.

It is easy to accidentally engage in role play, which will make your transcript weirder without improving the actual outcome. Use this method to find fields and techniques. Don't ask the model to write a short seller, ask it to do what a short seller does to research their position.

All the ways in which I describe ambition require the model to have a computer with the ability to install software and access to all kinds of relevant data (either using MCP connectors or other means). Plenty of tools make this possible and I recommend that you use one.

We have been here before

That's my concrete advice for how to be more ambitious. If you're willing to stick around for a minute, I'd like to tell you why I think this is even necessary in the first place.

We humans have always been held back by what we believe possible - and what's worse, we usually trust ourselves only in very narrow disciplines. Michelangelo famously didn't want to paint the Sistine Chapel - you see, he was a sculptor, not a painter.

While researching the problem of ambition, I learned that psychology has a name for this: self-efficacy. In short, people with high self-efficacy believe in their own abilities more, and in turn actually accomplish more. There's evidence that self-efficacy can be hacked: Locke and Latham pored over 35 years of goal-setting research to confidently proclaim that humans who are given specific hard goals perform better than those who are simply told to try their best. Stripe founder Patrick Collison maintains a "Fast" list, chronicling the many things humans have accomplished when given both audacious goals and tight deadlines. Some hacks that are well documented:

  • Accomplish hard things by believing them to be possible: In 1939, George Dantzig showed up late to a statistics class at Berkeley, copied two problems off the blackboard, and handed in his solutions thinking they were homework. They were two unsolved problems in statistics.

  • Accomplish hard things after watching somebody else do them: For decades, humans chased the four-minute mile. Roger Bannister broke it in May 1954 but remained the only human to do so for a mere 46 days. If you're into running, we're watching something similar happen to the two-hour marathon: Sabastian Sawe ran the first official sub-two-hour marathon in London this April. In the same race, the runner-up also broke the two-hour mark.

  • Accomplish hard things by giving yourself permission to fail: Apparently everyone's favorite business book is The Inner Game of Tennis, which I discovered just last week. The book's core thesis is that you're not playing your best game of tennis if you're trying really, really hard. We engineers might remember Linux, announced as "just a hobby, won't be big and professional like gnu".

Self-efficacy isn't a new problem, but it is a more powerful one now. In a world where our abilities make leaps every couple of weeks, we need techniques that allow us to do things we didn't really consider or believe possible. If you find more techniques that allow you to be more ambitious, please let me know!

The image for this post is by Francisco Goya, “Modo de volar” (A Way of Flying)

— END —

Felix Rieseberg Im an EM at Anthropic, where Ive so far lead engineering for Claude.ai, Cowork, and Claude Code Desktop. Im co-maintainer of Electron and various other open source projects. Ive previously worked on and for Notion, Stripe, Slack, and Microsoft.

MORE POSTS Older Made by mechanical means All posts

INDEXNEWEROLDERHOME

📋 讨论归档

讨论进行中…