关于 AI、产品、知识与组织的思考。

53 篇文章
文章地图
2026-11-07

《Malorean Corp.》成书了

当每个人都能做原本属于别的职业的事,谁来为什么负责?这本刚出版的书给出的答案是 Garant(担责人):每个领域都有一个人为它的一致性负责,但不必事事审批,也不必事事亲为。读完你会得到一种办法:既保住自主权,又不让它在混乱中消散,也不被层层审批压垮。

2026-10-29

Ontology-Driven Design:如果本体成为 AI 生成代码的安全网?

当 AI 代理编写代码时,测试能通过,业务却可能悄悄漂移:同一个概念出现两个名字,某条规则被违背,一个已被弃用的观念死灰复燃。从代码中提取本体,在每次开发前先与意图比对,再与产出的代码比对,就能让这些漂移显形,并把有意的演进与错误区分开来。

2026-10-28

抹平相互矛盾的客户反馈,是一种产品失误

“在简化的同时提高数据可靠性”这类综述能让所有人点头,却让真正的产品问题消失了。把每条客户反馈原样保存,一张卡片只写一个观点,再把张力本身写成一个独立的观点,一个令人尴尬的矛盾就变成了一个设计问题。原本天然倾向于抹平的 AI,也由此成为防止抹平的一道护栏。

2026-10-28

从零开始,还是照抄:两种丢掉已有知识的方式

第十次回答同一类安全问卷,既不想从头再来,也不想照抄一个错误的答案。把事实按主题整理,并注明来源、日期和适用边界,一份一百一十五个问题的问卷十分钟就能准备好百分之八十到九十,剩下的缺口则变成一个发给对口团队的具体问题。

2026-10-28

没有人知道如何重新接手一个案子

每次开会都要花二十分钟才能让大家回到同一起点,有人一走交接就丢三落四:一个共同跟进的案子,靠重读邮件是接不回来的。一张在工作过程中随时维护的接手页——进展到哪里、哪些地方不牢靠、只写一个下一个行动——能让每个人,也能让 AI,在一分钟内从同一个位置重新出发。

2026-10-28

交付物留住了结论,却丢掉了“为什么”

一条在某个客户那里奏效的建议,到了下一个客户那里可能就是错的,而报告不会告诉你原因。把每项决策连同备选方案、理由和成立条件一起留住,再从中提炼不含客户数据的笔记,事务所就能复用学到的东西而不背叛任何人,并在第一天就看出一个好答案何时已经不再适用。

2026-10-20

`Contexte` 现在是一本书了

你和 AI 的对话只留下结论,丢掉了理由:你排除过什么,哪条异议还悬而未决,你当初为什么那样决定。这本新书讲的,就是为解决这个问题而搭建的空间——一个主题一个空间,AI 能在里面找到一个想法的出处,并让工作从上次停下的地方继续。

2026-10-14

不用 AI 的 Graphify:零 token 把 wiki 变成可导航的图谱

一千条笔记彼此链接,却看不到全貌:哪些想法是核心,哪些是孤立的,两个相距甚远的概念如何相遇,都无从得知。如果 wiki 里每条链接的含义早已写在笔记中,一个 160 行的 Python 脚本就足以从中生成完整的图谱,再交给 Graphify 探索。整个过程只需 0.2 秒,不花一个 token。

2026-10-04

Stateful Corpus Research Loop:别再收集来源了

关键在于知道下一步该检索什么,更在于知道何时停下。与其事先定下 30 或 50 个来源,不如从一个小而精选的语料库出发,从中提炼出结构化知识,再让它已经知道的和尚不知道的东西决定下一次检索。结果是:冗余更少,盲区可见,停止标准也可以检验。

2026-09-23

那一天,AI 教会了我一件关于我自己系统的事

一份文档本身什么都不产出,却约束着之后产出的一切,它该放在哪里?常用的标准——变化的快慢——不够用。取而代之的是一个按作用划分的标准:这份文件是在产出结果、保存知识、描述当前状态,还是在约束未来的行动?这个标准适用于架构未曾预想的任何对象。

2026-09-21

我去查了查,“Contexte”里到底有什么是我真正发明的

花了几个月搭建的方法,会不会早已以别的名字存在于别处?查证的做法可以归结为几步:把想法拆解成机制,检索功能而不是词汇,走出本学科和英语世界,并事先规定一项新发现必须改变什么才算数。每一轮检索,可以宣称为原创的部分都在缩小,而方法本身的定位反而更清楚了。

2026-09-20

支柱卡片:让记忆保持鲜活,才能继续思考

借助 AI 快速产出,最后会挤掉用来思考的时间。这里给出一套具体的做法来守住它:一个有限的卡片核心,靠间隔重复保持鲜活,再摊在桌上,拿新想法与自己的信念对照检验。读完你会知道:什么样的想法才配进卡组,复习的节奏怎么定,以及第一步从哪里开始。

2026-09-20

直接在 ChatGPT 里运行我的 Contexte

不必依赖某处一台开着的 Mac,也能用 Contexte 工作。整套方案只有三块:ChatGPT 负责推理,项目指令定义引擎,Google Sheets 负责记忆。实际效果是:在展会上边走边口述一段对话,持久化,摄入,然后在离场前问系统还有哪些问题没问。

2026-09-13

企业级 SaaS 正在改变:当大型集团想要收回控制权

一个大客户卡住了你的单子:他要求数据放在自己的 Azure 里,用他的 IAM、他的密钥、他的 LLM 合同。本文给出一套筛选这类要求的框架:哪些真正承载你的价值,哪些只是让产品买得进来,以及客户的控制权和你推动产品演进的能力之间,边界该划在哪里。

2026-09-12

"Persist that": getting a piece of information out of ChatGPT and into permanent storage

Five days later, you rarely find the one paragraph that mattered in a long ChatGPT conversation. This is the story of how I got a precise fragment out of the flow on demand, using nothing but a custom instruction and a Google Sheets outbox — a blocked API, a workaround that failed, then a solution that went from 50 seconds to two, with a systematic check after every write.

2026-09-07

How I write my articles by talking to my phone

Your best article ideas die between the moment they turn up and the moment you sit down to write. Here is the setup that catches them while they are still warm: the voice to empty your head without putting anything in order, a session source to keep the raw material raw, AI to surface the gaps in ten seconds rather than in two days.

2026-09-04

The deliverable is no longer the asset

Every new format feels like starting over: you reopen the last document, copy, rephrase, and two versions slowly drift apart. The cause isn't a missing tool, it's treating a deliverable as your source of truth. How to install a reusable layer between your sources and your documents, and what AI gains from it.

2026-09-02

AI doesn't replace judgement, it speeds up how an idea matures

You can feel that a piece of writing or a decision isn't right yet, without being able to say why. That's where AI earns its place: it turns a vague unease into concrete handles — a blind spot, a shaky assumption, an objection you never dealt with. Which questions get you pushback rather than output, and why this makes human judgement more important, not less.

2026-08-11

I wrote a product's ontology. Three times, I thought I was done.

Your team rebuilds the same product model for every question, every bug, every tender. Here is how to write it down once: several dozen objects, a few hundred rules, under six hours. And above all how to know you have finished — the three stopping criteria that look like proof of completeness and prove nothing, plus the one that holds.

2026-08-03

The PO is not a profession, it's a function — and AI changes nothing about that, it just speeds it up

Hiring a PO to hold the backlog and sit between the business and the developers? You are optimising a circuit that is already obsolete. This piece takes apart the split between the people who do the thinking about the product and the people who execute it: why it degrades quality, why a backlog cannot support a profession, and why AI removes the last excuse for living with it.

2026-08-01

PM, developers and AI: roles blur, responsibilities remain

AI raises the capacity to build, not the capacity to decide. Developers move up towards product; PMs move down towards building. The shift blurs roles without dissolving responsibilities. A full pass over what AI makes ordinary in product management, what stays hard, and a catalogue of emerging roles with the real risk attached to each.

2026-07-28

Four days of vibe coding in the shoes of a rusty PM

Four days building a real product with Claude Code — 180 commits, starting from nothing. What it shows: the speed that goes to your head, the traps the AI creates on its own, what tests change when you haven't written the code yourself, and how far a PM can go without stepping outside their remit. A field report, not a doctrine.

2026-07-23

AI shouldn't just let us produce ten times more. It should also force us to understand ten times better what we produce

Producing code faster is worth nothing if nobody understands what got built. AI automates precisely the tasks that used to train junior developers, and an error that lands too late no longer teaches anything once the system has grown too big to analyse. A few concrete anchors for staying in control: clear contracts between APIs, tests tied to the functional need, rebuildability as a quality bar rather than raw delivery speed.

2026-07-20

Session sources: what your documents miss

Your documents never say everything. The fine-grained understanding of what happens on the ground stays in the PM's head. Session sources open a channel for pouring that informal knowledge into an AI Context — a hallway remark, a nuance picked up from a customer, a tension just starting to show — and turn dozens of micro-signals into notes, assumptions and better-framed decisions.

2026-07-08

Information systems are not going away. Their shape will change.

Your tools file information away neatly, but they never build up what that information teaches you. This piece explains why the future of the information system is not a better AI-assisted form, but a shift towards `Contexts`: portable, governed units of memory that connect sources, decisions and reasoning instead of merely storing them.

2026-07-06

The product question should start from the source code

Do you lose half an hour in a test environment just to answer a product question on Slack? That hidden cost disappears once AI can query the source code, the real truth about how the product behaves. Answers grounded in evidence — functions, tests, commits — rather than in documentation that has fallen behind. Provided the code speaks the language of the business.

2026-07-01

Under the bonnet of my context engine: how an AI remembers a mission

Do you lose the thread the moment you reopen a conversation with an AI about something that has been running for weeks? This article lifts the lid on a structured memory system: the difference between the goal and the attention of the moment, the directives, the offline queue, the three memories (log, synthesis, checkpoints) and the cycle of a session. Enough to see how an AI can genuinely remember a mission.

2026-06-17

A product doesn't enter a country through language, but through everyday practice

You think a market opens up once your product is translated? Brazil says otherwise. Through the most mundane scene there is — paying at the till — you'll see why everyday credit, a tax ID number, Pix and a visible informal economy form a cultural infrastructure no translation reproduces. Enough to avoid an expensive strategic mistake abroad.

2026-06-16

Flash Card

An excellent note base isn't enough if the ideas never get absorbed. A flash card isn't a small atomic note: it's a personal pillar of thought, recognised in the heat of the moment by its resonance, tested over time by experience and by new sources. Rare by definition, it holds up what you really want to carry in you.

2026-06-16

The art of capture

We pile up tabs, links and quick notes — and never read them again. The problem isn't the amount of information, it's that nothing is ever decided about what it becomes. This article separates capturing from building capital, and offers a simple rhythm for turning your captures into knowledge without drowning in them.

2026-06-16

What is a glossary entry?

The same words travel between product, marketing, sales and leadership, and nobody gives them quite the same meaning. A glossary isn't a collection of textbook definitions: it's a tool for collective precision. It pins down the terms that carry decisions, protects the links between notes, and cuts the cost of conversations where everyone defends a different idea with the same vocabulary.

2026-06-16

What is an atomic note?

"One idea per note" is the formula everyone knows, and it sends most people in the wrong direction. An atomic note is not a short note. It is a note with a single conceptual centre of gravity. Once you see that difference, it changes how you capture, rephrase, title and link. And in the end, how you write.

2026-06-16

What Is a Thematic Note?

A thematic note is the link between a note system and writing. It doesn't pile up everything that touches a subject — it picks an angle and brings several atomic notes together to make a pattern, a tension or a line of argument visible. Not a folder, not an encyclopedia: a temporarily stable view onto a network of thought.

2026-06-08

The PM as Architect of Context

Most organisations know how to produce documents. They don't know how to bank the reasoning behind them. On every decision they start again almost from scratch. The PM as Context architect turns scattered customer knowledge — transcripts, objections, past calls, weak signals — into a reusable asset. The result: faster, sturdier decisions, and an advantage competitors cannot copy overnight.

2026-06-03

Quality belongs to those who ship

A zero-bug policy shrinks the stock of known defects. But if the team fixes them faster without changing how it builds, it stays stuck in a repair loop. The real issue lies upstream: giving ownership to those who ship, refusing overly vague specs, testing earlier, giving QA a quality-policy role rather than a catch-up one. Quality can't be delegated after the fact — it belongs to those who ship.

2026-06-03

The Backlog Is Not a Dumping Ground: It's a Tool for Action

In many organizations, the backlog has become a clean trash can: you throw everything in so you don't forget. A few months later, 500 lines nobody understands anymore. A backlog shouldn't be a repository of ideas, a database of customer signals, or the graveyard of untriaged bugs. It exists to organize information mature enough for several people to work on it together. It comes into play when the work stops being individual.

2026-06-03

Product Decision Record: tracing the product choices that shape the company

Certain product decisions come up everywhere: in roadmap reviews, with sales, in specs, with every big client. Without a record, the company keeps re-deciding them, often with less context. The PDR — inspired by technical ADRs — documents these cross-cutting decision rules. Rare, short, immutable: it doesn't say what to build, it says why the company chose a rule. It removes dozens of pointless discussions.

2026-06-03

The NNL Roadmap: Aligning Without Scattering

A product team can pile up studies, workshops, and analyses without the company ever knowing what's a priority. The NNL roadmap answers this problem: not by scheduling dates, but by making commitments, serious directions, and possibilities visible — and above all, what won't be done. Its strength lies in the capacity constraint and in the implicit No.

2026-06-03

Zero bug: let's stop managing stockpiles of defects

The real problem with bugs isn't that they exist, it's the organization that grows used to keeping them. We prioritize them, we push them back, we review them in committees — until a customer crisis hits. A zero-bug policy aims at something else: zero known bugs left undecided. Either we fix it, or we accept it isn't a defect. What this approach refuses is the third state: knowing a bug exists and keeping it alive in a list for later.

2026-06-02

The Tools of Organizational Coherence

Your teams leave the meeting aligned—and each walks away with a different version of the decision. The problem isn't a lack of meetings, it's the absence of shared objects. Glossary, one-pager, PR/FAQ, ranked metrics, decision log, escalation rules: these tools don't produce alignment, they force disagreements to become visible before they turn costly.

2026-06-02

Why Organizations Prefer Soft Decisions

You leave a meeting with an action plan, yet nothing has truly been settled. This is what I call a soft decision: a non-decision dressed up as consensus. Organizations often prefer it to a genuine judgment call because it preserves the appearance of agreement without making the loss visible. This piece explores why the mechanism is rational in the short term, and why it costs so much at the execution stage.

2026-06-02

Competitive intelligence: copying your competitors is not a strategy

Watching what competitors ship means reading outputs without understanding the reasoning behind them. Useful intelligence starts from the target customer: their needs, their decision criteria, their alternatives. Competitors only become interesting because they reveal how a market responds to those needs. Copying a move without understanding the logic behind it means reproducing an answer without ever having asked the right question.

2026-05-25

AI Wiki: why I built a knowledge base maintained by an AI

Classic RAG starts from scratch with every question. By building an AI Wiki — a persistent wiki maintained by an LLM — you capitalize on domain knowledge once and for all. Atomic notes, links between concepts, contradiction detection: knowledge is compiled, not rediscovered. Lessons learned from 950 notes extracted from industrial maintenance sources, with calibration insights and concrete use cases.

2026-05-03

The second brain is a dead end for product management

Managing product opportunities inside a borderless network of notes means losing control: errors propagate silently, the AI context window blows up, and contradictions and decisions become impossible to find. The bounded context — borrowed from DDD — offers a better approach: each opportunity in its own bounded, auditable space, with a structured memory.

2026-04-24

One file, a few directives, and Claude does the rest — how I structured 500 emails effortlessly

Drowning in email and losing track of the threads in flight? By handing the structuring over to Claude — one file, a few directives, then a gradual split — I turned 500 building-management emails into a queryable knowledge base. Seven steps, from the vague directive to a multi-file system, without ever needing a complex framework.

2026-04-21

From the Single File to a System of Contexts: Why an LLM's Memory Won't Fit in One Document

2026-04-13

Code Centric

Documentation never keeps up with the code. We know it, we accept it, and yet it costs us time with every change. AI shifts the equation: when the code is clean, it becomes the source of truth from which the other artifacts—specs, changelogs, support documentation—can be regenerated. This code-centric approach demands a minimum of technical literacy, but it opens a real lever for the product managers who master it.

2026-04-01

Why a single classification isn't enough to structure customer feedback

Too many insight systems fail because they classify everything along a single axis: features, or business stakes. Tactical feedback gets lost, strategic signals evaporate. A 4-level framework — business stake, use case, product capability, local friction point — lets you capture everything without a Rube Goldberg machine, and ties each signal to the right level of decision.

2026-03-25

Why You Should Audit Your Skills: Master the Hidden Token Limits

A wasteful skill locks you out after 3-4 runs instead of 8-10. Discover the 5 silent traps that saturate your quota, how to audit in 3 simple phases, and how to gain 40% more capacity without changing your model. Productivity regained.

2026-03-23

In 8 days, I understood that the Product Manager job was about to change completely

Your days as a PM are devoured by tasks that have nothing strategic about them. In eight days on Claude Code, I automated my competitive intelligence — gathering, organizing, updating. Time reclaimed to analyze, understand, decide. This field report shows concretely why the center of gravity of the Product Manager job is shifting: fewer intermediate artifacts, more real strategy.

2026-03-23

In software, the edge will no longer be technology. It will be understanding the context.

Producing software is no longer rare. The cloud commoditized it, and AI will accelerate the trend further. When everyone can build fast and well, churning out features by the yard is no longer enough. The real competitive edge shifts toward understanding the context: the business, the users, the customers, the competition, the standards. Understanding better than others, before building.

2026-03-22

Adding session memory, like OpenClaw

No more re-explaining the context at every session. Claude automatically retains what you did, the decisions made, the problems encountered — per project, inside your git repo, with no database. Portable, versionable, under $1/month. Here's how to set this system up in less than an hour.