Knowledge & data
AI knowledge system
Every business holds knowledge that lives in a single head. How to price the tricky project. Why one client needs special handling. What really matters during onboarding. When that head is on holiday, the question hangs in the air. And even the knowledge that is documented somewhere costs time every day: searching, asking, waiting for someone to reply.
AI knowledge system
01AI knowledge system
A knowledge base answers questions. A knowledge system is more: the place where notes, projects, standards and processes actually live – structured so that AI can search, summarise, take notes and run defined routines in it. Not another tool alongside the work, but the system the work happens in. The entry point is deliberately small: one bounded area of knowledge is structured and made queryable – staff find standards and answers instead of interrupting colleagues. From there the system grows: first the AI only reads, then it drafts and logs with sign-off, eventually it takes over recurring routines. I work this way myself every day – a large part of my daily business now runs through exactly such a system, with AI as a colleague inside it.
Up front: I won't promise a "second brain in one go" – start too big and it fails on upkeep. Whether your business already carries such a system today or Tier 1 is enough for now is nothing to guess at: a small AI audit beforehand tells you whether the build pays off – the check first, then the system. Sensitive data (clients, patients, staff) stays out or gets its own data-protection concept. The AI never writes into business-critical areas unchecked. And the system only lives if the team carries it – which is why "bring the team along" is a mandatory part of the project here, not an option.
02Demo
The knowledge system of the fictional “Brandl Carpentry” – on the left the order in which the company's knowledge lives, on the right three everyday moments with it. Click through and watch where it lights up in the tree.
The architecture behind it has two layers: at the bottom the vault – structured folders and notes that belong to the business. Above it the AI layer with documented rules: it reads only at first, then writes with approval, and sensitive areas stay locked. Calendar and tools dock on, routines run within clear limits.
The most honest proof of this model: this website came about exactly this way – I work in such a system myself every day, with AI as an employee inside it.
03How it works
- 01
Take stock: where does your knowledge live?
Heads, PDFs, email threads, specialist systems, notes stuck to the monitor. Plus: which processes repeat, who works how, where does knowledge fall away when someone leaves?
- 02
Build the structure, small and documented
A clear ordering and note system with simple, written-down rules – the foundation everything else sits on. Boring, decisive, and the step everyone wants to skip.
- 03
Open up and sort the knowledge
Existing material moves in – curated, not copied. For larger legacy holdings, Intelligent digitisation does the groundwork.
- 04
Connect the AI: reading first, then writing
First the AI may only answer questions, find, summarise. Once that holds, writing with sign-off is added: drafts, minutes, entries. Rights and limits are set down as a documented work instruction.
- 05
Add routines – step by step
Recurring reports, logs, checklists, reminders: the AI takes over defined tasks within clear limits. Not all at once – each routine only once the previous one runs.
04Stages
The entry point is Tier 1, the team knowledge base – that's exactly where every project begins.
Entry: the team knowledge base
One bounded area of knowledge is structured and made queryable – standards, processes, answers. Staff look things up instead of interrupting.
Extended: the AI works along
Summaries, drafts, minutes, log upkeep – the AI writes along in the system, every change with sign-off.
Full build: the agentic system
Defined routines run on their own – reports, reminders, recurring tasks. The knowledge system becomes the work system.
05Technical implementation
- The knowledge layer is yours: a Markdown vault (e.g. Obsidian), Notion or a comparable system – content without vendor lock-in.
- The AI layer is an assistant with access to the vault (e.g. Claude Code as the agentic layer) – with documented rules: which areas, which rights, which routines.
- Tiered rights logic: reading first, writing only in defined areas with sign-off, sensitive areas excluded.
- The core isn't the tool but the combination: structured order + clear rules + tiered sign-offs.
- Upkeep routines are set up alongside – a knowledge system without a maintenance process ages like any wiki.
- Typical sources: standards, checklists, copy modules, project and meeting notes, onboarding material, project learnings.
for context · This shows how something like this usually gets built – not how it has to be built for you. What actually fits is worked out in the project: around your existing setup, your team, and what stays easy to maintain.