Maggie Appleton 写设计、人类学和智能体界面时，习惯先画清楚隐喻，再谈工具。把「Visually Workshopping the AWS Cloud」整理成可阅读的中文笔记：问题在哪、界面默认会把人带去哪、落地时该改什么。原站导航、评论回响和广告已去掉。
The Process
I’ve made plenty of illustrated notes for egghead courses. On everything from Building Custom React Hooks to Fixing Common Git Mistakes to The Basics of Rust
They’ve always been artefacts made in post-production. I draw them after the course has already been planned, and the videos are all recorded. Over the last month I’ve been experimenting with a more collaborative approach.
The Whole Series
Rather than treating our illustrations as a shiny add-on made by one person, I wanted to directly involve our course instructors and figure out how to directly pair their expertise with my visual thinking skills . So I started running visual workshopping sessions.
We hop on a Zoom call and open up a shared canvas in Miro or Excalidraw . They’re both great and do the trick for the task at hand. I recommend playing with both! We pick out a few of the toughest concepts, and begin making extremely rough diagrams. Experimental ugliness is the whole point.
Collaborative Minds > Beginners Mind
We just did our first trial of this with Tomasz Lakomy ’s new course on AWS Cloud Development .
Here’s Tomasz thinking very hard about API Gateways and Lambda Functions
落地时建议先做的 5 件事
- 先写清这个工具在强化思考还是在替人思考。
- 默认交互不要只会谄媚：该追问、该给反例、该要求证据。
- 智能体界面要露出推理步骤，而不是只给一个光滑答案。
- 知识发布用可生长的笔记，而不是一次性营销长文。
- 改产品前先画隐喻：用户以为自己在做什么，系统实际在做什么。
和智能体产品怎么接
龙虾PRO做 OpenClaw 落地时，最该从这类笔记里拿走的是「别把聊天框当唯一界面」。数字员工要能追问、能验证、能把过程摊开，而不是只负责说好话。
本文侧重智能体生命周期与技能边界。若你要动手验证，优先用 Playground 做小任务压测；权限与托管再单独规划，避免「先装一堆再治理」。 相关：Playground · 创建智能体 · 学习中心
常见问题 FAQ
什么是知识与智能体怎么？
「知识与智能体怎么」可概括为：I’ve made plenty of illustrated notes for egghead courses. On everything from Building Custom React Hooks to Fixing Common Git Mistakes to The Basics of Rust 本文从定义、方法与实践要点展开说明。
为什么要关注知识与智能体怎么？
关注知识与智能体怎么，是因为它直接影响效率、风险与可复制性。文中指出：I’ve made plenty of illustrated notes for egghead courses. On everything from Building Custom React Hooks to Fixing Common Git Mistakes to The Basics of Rust
如何落地知识与智能体怎么？有哪些关键步骤？
建议按以下路径推进知识与智能体怎么：1) 先写清这个工具在强化思考还是在替人思考。；2) 默认交互不要只会谄媚：该追问、该给反例、该要求证据。；3) 智能体界面要露出推理步骤，而不是只给一个光滑答案。；4) 知识发布用可生长的笔记，而不是一次性营销长文。；5) 改产品前先画隐喻：用户以为自己在做什么，系统实际在做什么。。细节见正文对应章节。
知识与智能体怎么适合哪些人或团队？
知识与智能体怎么更适合：产品/技术负责人、运营与增长团队、需要落地智能体或自动化的中小团队、关注「AI智能系统」方向的读者。若你只需要单次聊天式问答，可先读概念；若要上生产，请重点看步骤、权限与风控相关段落。
关于「The Process」，本文给出了什么结论？
在「The Process」部分，要点是：lways been artefacts made in post-production. I draw them after the course has already been planned, and the videos are all recorded. Over the last month I’ve been experimenting with a more collaborative approach. The Wh
关于「The Whole Series」，本文给出了什么结论？
在「The Whole Series」部分，要点是：pertise with my visual thinking skills . So I started running visual workshopping sessions. We hop on a Zoom call and open up a shared canvas in Miro or Excalidraw . They’re both great and do the trick for the task at ha