Notes on building

Wide illustrated banner showing an AI assistant assembling an answer from multiple product documentation cards, with one clearly labeled startup product card being selected and highlighted among others, representing how to make your product show up in AI answers

Oct 6, 2026 · Building productsHow to make your product show up in AI answers怎麼讓你的產品出現在 AI 的回答裡→A concise playbook for getting your product to surface inside AI generated answers, by treating LLMs as readers of your public knowledge and optimizing what they can safely quote.把 AI 想成會讀你網站的超級使用者，從內容、結構到行為，一步步提高產品被 AI 答案引用的機率。

Wide banner showing a stylized brain made of interconnected folders, chats, and diagrams, glowing in orange, with small AI agents moving between them to suggest a shared company brain that connects tools, people, and knowledge in one place.

Oct 2, 2026 · Building productsWhy AI needs a company brainAI 為什麼需要一顆「公司大腦」→Most teams plug AI into their tools without giving it a shared company brain. Here is why that limits impact, and what to prepare before you wire AI into how your team really works.多數團隊把 AI 接進工具，卻沒有給它一顆「公司大腦」。這篇談為什麼那樣很難有影響力，以及在把 AI 接入日常工作前，你該先準備什麼。

A product dashboard viewed from above where one side is a human hand and the other is a subtle robotic arm, both clicking the same interface, illustrating that your next user is an AI agent using your product like a human would

Sep 29, 2026 · Building productsYour next user is an AI agent你的下一個使用者，是 AI agent→Why the next real user of your product may be an AI agent: what changes for product thinking, docs, and interfaces when software starts using your software.當 AI agent 開始「用」你的產品，產品思維、文件和介面會跟著變形。這篇整理你現在就該調整的幾件事。

Abstract dashboard interface where an old, greyed out SaaS product is overlapped and partially replaced by a bright, orange highlighted AI driven interface, illustrating how every SaaS category is suddenly up for grabs for new products and challengers

Sep 24, 2026 · Product planningWhy every SaaS category is suddenly up for grabs為什麼每個 SaaS 品類突然都能被挑戰→SaaS used to feel locked up by incumbents. Now AI, infra changes, and shifting buyer expectations are reopening almost every category at once.SaaS 一度像是被巨頭封印，但 AI、基礎設施與買方心態的改變，正同時打開幾乎所有品類的缺口。

Wide illustration of a small product team in a studio surrounded by laptops, whiteboards, and flowing data streams, showing an AI-native service company that designs, builds, and operates custom AI workflows for clients instead of only shipping static software projects.

Sep 22, 2026 · Building productsWhat is an AI-native service company?什麼是 AI-native 服務公司？→A practical look at what makes a service company truly AI-native: how it works, hires, prices, and learns differently from traditional agencies and software shops.從實務角度拆解 AI-native 服務公司的關鍵：和傳統接案、軟體公司在流程、人才、定價與學習模式上，有哪些根本不同。

A founder at a desk with sketchpaper, sticky notes and a laptop, comparing different price tags for a new digital product, illustrating how to price your first product without fear or guesswork

Sep 17, 2026 · Product planningPricing your first product without fear第一個產品該怎麼定價：不要被價格嚇到→A practical way for first time founders to price their product without overthinking it or racing to the bottom.給第一次做產品的你，用簡單結構搞定定價，不再瞎猜也不再自己打價格戰。

A simple product roadmap drawn in reverse on a whiteboard, starting from a bold impact statement on the right and stepping backward to smaller milestones on the left, illustrating how to plan backwards from the impact you want

Sep 15, 2026 · Product planningPlan backwards from the impact you want以終為始：從你要的影響力往回規劃→Stop planning features from a blank page. Start from the specific impact you want, then work backwards into growth, MVP scope, and daily decisions.別再從功能清單開始規劃，先說清楚你要創造的影響力，再往回推導成成長策略、MVP 範圍與每天的選擇。

An early stage founder sits at a small desk at night, surrounded by instant noodle cups, a laptop with product mockups, and a simple profit chart just crossing into the positive, symbolizing the quiet but critical milestone of reaching ramen profitable for a startup

Sep 10, 2026 · Building productsRamen profitable: the underrated milestone泡麵獲利：被低估的里程碑→Why reaching ramen profitability quietly changes your startup from a fragile idea into a durable, default-alive product company.為什麼「泡麵獲利」這個小小門檻，會悄悄把新創從易碎的點子，變成能活下去的產品公司。

Wide orange tinted illustration of a founder quietly observing people at work in a café, sketching notes about their workflow on a laptop, symbolizing how good product ideas come from noticing real problems instead of random brainstorming

Sep 7, 2026 · Building productsWhere good product ideas actually come from好的產品點子到底從哪裡來→A short field guide to where strong product ideas really start: from lived problems, organic insight, and a higher quality bar than “this might work.”一份精簡的小指南：好的產品點子不是靠亂猜，而是從真實問題、有機洞察，和比「好像可以」更高的標準開始。

Founder at a desk manually emailing early users and checking notes on a whiteboard, illustrating a do things that do not scale approach to finding product market fit

Sep 3, 2026 · Building productsThe do-things-that-don't-scale playbook不能規模化的事，怎麼做才對→A short playbook for doing the unscalable work that actually moves early products: from Paul Graham’s advice to Superhuman’s PMF engine and The Mom Test.新產品一開始就是一連串「不能規模化的事」。從 Paul Graham、Superhuman 到 The Mom Test，整理一份好好去做的實用清單。

Minimal orange and white illustration of a small product rocket launching repeatedly in short cycles, visually reinforcing the idea of launching early and then keeping a steady cadence of incremental product launches

Sep 1, 2026 · Building productsLaunch early, then keep launching早點上線，然後不停地上線→Why small, frequent launches beat one big reveal, and how to plan an MVP that can ship fast then keep evolving in public.為什麼小步快上線，比憋一個大版本更有效，以及怎麼規劃 MVP：先丟出去，再在真實世界裡一路長大。

A small product team in a dim studio, surrounded by glowing screens and sticky notes, arguing over a roadmap while a quiet crack spreads through the floor beneath them, symbolizing subtle mistakes that slowly kill new products before anyone notices

Aug 28, 2026 · Building productsThe mistakes that quietly kill new products悄悄殺死新產品的那些錯誤→From Paul Graham and YC: the subtle early stage mistakes that slowly kill new products, plus a short checklist to keep you honest.從 Paul Graham 和 YC 的建議，看那些在早期默默累積，最後把新產品拖垮的錯誤，外加一份簡短自我檢查清單。

A founder sits at a desk at night with a laptop and a simple financial spreadsheet, comparing a rising and a falling line chart to decide whether their side project is default alive or default dead under current assumptions

Aug 24, 2026 · Product planningIs your project default alive or default dead?你的專案是「預設活著」還是「預設死亡」？→A quick way to see if your project can survive on its current path, using Paul Graham's ideas on default alive, reasons startups die and ramen profitability.用 Paul Graham 的幾篇文章，快速檢查你的專案：照現在這樣走下去，是活得下去，還是註定餓死？

Aug 23, 2026 · Building productsYour UX process is a 1960s ad agency relic, and AI just exposed it你的 UX 流程其實是廣告黃金年代的遺產，AI 只是把它戳破了→Pitch decks, star designers, billable deliverables: UX Collective argues most in-house UX teams still run on Mad Men era ad agency habits. As AI collapses the cost of producing polished mockups, the gap between looking good and shipping outcomes is getting impossible to hide.提案會議、明星設計師、按交付物計費：UX Collective 一篇文章指出，多數企業內部 UX 團隊骨子裡還在用廣告黃金年代留下的一套運作習慣。當 AI 讓做出漂亮提案的成本趨近於零，這套習慣跟真正的業務結果之間的落差，已經藏不住了。

Wide orange banner showing a small product icon snapping into a matching market puzzle piece, surrounded by fading outlines of other pieces, to illustrate that product market fit is the only thing that matters for a startup

Aug 20, 2026 · Building productsProduct market fit is the only thing that matters在 Product Market Fit 之前，其他都不重要→A concise tour of why product market fit dominates everything else, what Marc Andreessen, Rahul Vohra and Paul Graham actually mean by it, and how to prepare before you chase it.從 Marc Andreessen、Rahul Vohra 到 Paul Graham：為什麼在找到 Product Market Fit 之前，其他幾乎都不重要，以及你在起跑前能先準備什麼。

Minimal orange gradient banner showing a small group of people with speech bubbles gradually converging into a single highlighted product icon, visualizing the idea of literally making something people want instead of what founders imagine

Aug 17, 2026 · Building productsMake something people want, literally把「做人們想要的東西」當真一次→YC and Paul Graham repeat this advice so often it becomes wallpaper. This piece unpacks what “make something people want” means in practice, and how it changes what you do this week.YC 和 Paul Graham 把「做人們想要的東西」講到變成背景噪音。這篇試著拆開看，變成這週可以真的去做的幾件小事。

Orange gradient banner with six feature blocks where only two survive, illustrating how to cut an MVP down to its core

Jul 8, 2026 · Product planningHow small should your MVP be?MVP 到底該多小？→Three classic answers from YC, Paul Graham and Steve Blank on scoping a first version that ships in weeks, not months.整理 YC、Paul Graham 和 Steve Blank 的三個經典答案：第一版就該小到讓你有點不好意思。

Orange gradient banner with a question bubble and an answer bubble, illustrating user interviews before building a product

Jul 8, 2026 · Building productsTalk to users before you build anything動手做之前，先跟使用者說話→What Paul Graham, The Mom Test and the Superhuman story teach about the cheapest research method there is: a real conversation.Paul Graham、The Mom Test 和 Superhuman 的故事，教你用最便宜的研究方法：一場真實的對話。

Notes on building.打造筆記