People forgive a blurry picture faster than bad sound. Here is how you sort out your online setup, with a backup in case something does fail.

An online meeting that starts with 'wait, I cannot hear you', you have probably experienced that. And then someone says their camera is not working. Perhaps you have even said it yourself. The meeting then starts with five minutes of technology, while everyone sits looking at a little black box.

In a training session on digital and hybrid meetings, we went over what your setup needs together. I also honestly said there that I no longer consider 'my camera is not working' an excuse. That sounds strict, and it is a bit. Because almost everything that goes wrong, you can check in advance.

What you build with AI must be readable by someone else later, human or agent. Ask your agent to write the documentation before you put something live.

You have built a little tool with your agent and it works. Your colleagues want to use it too. And then comes the question nobody likes to hear: how do I install that? You still roughly know it yourself, because you were there. But in three months you might not know it anymore either. And whoever continues working on it after you, certainly will not know.

In a vibe coding training I talked about an organisation with old software. They wanted to work with it more easily using AI, and I built a few tools for them. Building was actually only half the work. Other people had to be able to read how those tools work later on. And by extension their agents too. So everything had to be documented.

Prompts to paste into Copilot next to your open document in Word, Excel or Outlook, so it does the work in your file and you no longer have to copy and paste.

You have Copilot open in Word, in Excel or next to an email in Outlook. The document is there, the chat is next to it, and the cursor is blinking. And then you do not really know what to ask. So you ask something you would also ask in the standalone chat. And you get another text that you have to copy and paste yourself.

This swipe file gives you short prompts that you paste into Copilot while your file is open. They are written for the moment when Copilot is allowed to work in your document itself. You choose one, you fill in what is in brackets, and you see what it does. Take fifteen minutes for it, with a document you had to edit anyway.

A meeting that waits for latecomers costs the time of those already sitting there. This is how you respect the clock, from the first to the last fifteen minutes.

You are ready for a ten o'clock meeting. At five past ten, someone is still missing. So everyone waits. People chat a bit and check an email. And at quarter past, it finally starts. You probably recognise this, and perhaps you even find it normal. I think almost everyone experiences it like this.

In a training course on meetings, I started punctually at the agreed time. Even if not everyone had been there, I would have started. I asked the group why I find that so important. The answer came immediately: otherwise all the others are just sitting around waiting. And that is a waste of time.

Writing everything out in advance or building in short loops? Both are possible, but those who only see the result at the end often get a kitchen they did not want.

Perhaps you have already read that with vibe coding you have to write everything out first. A complete specification, right down to the last screen. Only then is your agent allowed to start. And perhaps you are sitting in front of an empty document, with the feeling that you do not know all of that yet. That is not a sign that you are not ready for it. A creative process simply does not work like that for everyone.

In a vibe coding training course we talked about two ways of working. One is called spec coding, from specification. The other works in loops, with progressive insight. And I honestly told the group that one is not better than the other. They suit a different kind of person.

In the standalone chat, Copilot mostly gives you text back. Open it in Word, Excel or Outlook, and it does the work itself right there in your document.

You ask Copilot in the chat to update your report. You get a neat text back. And then you have to paste it into your document yourself, piece by piece. You might wonder what that AI assistant is actually for, if you are still doing the cutting and pasting.

That question also came up in a Copilot workshop. The Copilot chat can normally do very little with your files. It does not put documents in folders and it does not change anything in your Word. But if you open Copilot in the app itself, that changes. There it just does what you need it to do.

Four prompts to tidy up your AI's memory: first testing what it still believes, then proposing, deciding and keeping the map small. You decide what can go.

Your AI has been given a memory, and for the first few weeks that was a relief. Meanwhile, it is full of all sorts of things. Logs of tasks, appointments that changed a long time ago, and the same note in three versions. Your AI reads all of that with equal conviction. It does not know which line is from last year. With these four prompts, you tidy up that memory in fifteen minutes. Your AI does the reading and proposes, and you decide what stays. It works with a folder of notes, a vault in Obsidian or a project in Claude or ChatGPT with files.

One card to hang next to your screen: four settings for your AI's thinking level, two questions to set the dial and two prompts for your own rule of thumb.

You probably have a button in your AI tool for how deeply it thinks. In Claude it is called thinking level or effort. In Copilot you choose between Auto, Quick response and Think deeper. And if you are like most people in my workshops, that button sits untouched on the default setting. Or it is set to the highest just to be sure. An email to your accountant on the highest level is actually a bit like a full inspection at your garage. And that because your windscreen wiper squeaks. That is very thorough, only you are waiting half a day for it.

You do not have to choose between an agent and a workflow. Let the agent create what is different every time, and a fixed workflow do what is always the same.

In a training course on automation, we ran a workflow at a fixed time. A participant asked me if that could also just be done with Claude or ChatGPT. And they suspected that you would use a lot more tokens then. Perhaps you have the same doubt, now that you hear everywhere that agents can do everything. Then building an ordinary workflow seems a bit old-fashioned.

The suspicion was correct. And I would add something else to it: you do not actually have to choose. The setups that run well in practice use both, each for the part they are good at.

Let your agents keep track of what they do, and their memory grows faster than you think. A regular clear-out keeps it small, a bit like dreaming.

You have given your AI a memory. That could be a folder with notes or a file with what it needs to know about you. Perhaps it is even a fully fledged second brain in Obsidian. For the first few weeks, that works very well. You have to explain less and your agent knows where to look. And then one day you notice that it is slower at finding its way. Or that it brings up something that has not been correct for a long time.

In a vibe coding training course, I showed how my own agents work with such a memory. And I had to honestly admit that I ran into this myself. At a certain point, my vault started to become very complex indeed.

A fifteen-minute worksheet for your review with your AI: first your own guess, then its answer, one choice, and two prompts to close the loop.

You have been working with your AI for a while, and it knows a fair bit of your work by now. In the article you read how to let it look back at what you did together. That answer sometimes hits hard. And then the temptation is great to read it, nod briefly and get on with your day. This worksheet helps you to do something with it. You need fifteen minutes for it, once a month or once a quarter. You end up with one thing you pick up and a note for your next review.

Three questions before you ask your AI anything

First take a piece of paper yourself. Write down a few words for each question, even before you paste the prompt from the article.

If your building project with AI gets stuck, pushing on a bit longer rarely helps. Stop in time, and talk your problem out loud, to a rubber duck, a colleague or your agent.

It is late, your app still does not do what you want, and you think: just one more prompt. Then another one. You might recognise this, if you have ever built something with AI. And the next morning you open your project and think: oh dear, what has happened here?

In a vibe coding training I talked about this with the participants, alongside the good practices I try to follow myself. Two of them seem to have nothing to do with technology at first glance. Yet they help you out of a knot more often than a better prompt.

Every workflow has points where it must choose which way to go. Here is how you determine for each fork whether a rule, an agent or you yourself decide.

You want to automate a piece of your work and you draw out the steps. Mail comes in, something reads that mail, and then something has to happen. Only, a choice has to be made at that moment too. Is this a real question or is it spam? Is this document an invoice or a contract? In a training course on automation, a participant asked me whether such a workflow is AI or not. A good question, because you probably also feel yourself that it has to be one or the other.

The answer is actually: both, and different for each fork. And the better you see those forks, the easier your workflow becomes to build.

Alongside your model is a second button: how deeply your AI thinks. Higher is not always better. Here is how you choose the right thinking level for each task.

Alongside the choice of your model, many AI tools have a second button. In one tool it is called 'think deeper', in another 'effort' or 'reasoning'. And perhaps you have had it on the highest setting for a while, because thinking more just sounds better. In a webinar about Claude, a question about this came up in the chat. Someone wrote that the combination of model and effort was never really clear.

I do understand that confusion. You have two buttons that both seem to have something to do with smartness. And no one explains what the second one does exactly. So you just set them both to the maximum, just to be sure.

If your agent asks for permission at every step, you will eventually click approve blindly. This is how you put control where it actually makes a difference.

You might know the feeling. You give your agent a task and it gets to work. And then it starts: may I read this file, may I run this command, may I edit this folder? The first few times you read every question carefully. After half an hour you just click through.

In a training course on automation I mentioned that I know plenty of people who go about their whole day like this. Laptop open, and then bam, bam, bam, clicking approve. And honestly, I understand that. But then you have a strange situation. The button that is supposed to protect you is the button you press the fastest.

Phrases you can say word for word to open a meeting, explore together, work with resistance and close. Plus a prompt for your notes.

You know you have to open first before you decide. And then you sit there, everyone looks at you, and the first sentence you say is your own solution anyway. That happens to experienced facilitators too, you know. Under pressure, your brain reaches for what it already knows. So it helps to have a few phrases ready for every moment of your meeting.

This swipe file gives you those phrases. They follow Sam Kaner's diamond of participation from the article: opening, exploring, narrowing down and closing. You can read through them in five minutes before your next meeting. And you just choose a few that suit you.

The same question, the same model, and yet a different answer? Then the app is deciding too. Here is how you read a refusal and know what to do with it.

You ask your AI something very ordinary and you get a polite refusal. You try the same thing in another tool, with the same model underneath, and there it works. That actually feels a bit random. In a vibe coding training, a participant ran into this too. She literally said: 'It is really strange. How is that possible?'

The answer is not in your prompt and not in the model either. It is in the app around it. And once you see that, you will read a refusal very differently. You then know where to look. And you also know when it is better to just stop trying.

No enquiry gets left behind and no follow-up gets forgotten. How AI helps you reply quickly and follow up on time, without it feeling cold to your clients.

You spent a whole day with a client. In the evening you look in your inbox and see an enquiry that came in early this morning. You reply immediately, but you already know. Anyone who asks a question in the morning and only gets a reply after dinner has already emailed another provider in the meantime.

I think that has little to do with carelessness. An enquiry that gets left behind is rarely a bad enquiry. Usually your day was just too full. And the fuller your days get, the more often this happens, and the less time you have to notice it.

The manual work that returns every week does not have to stay with you. What AI can take over from your emails, reports and quotes, and how you start small.

You are sitting at the kitchen table on Sunday evening with your laptop open, when you actually wanted to be on the sofa already. Quotes still need to go out. Three clients are waiting for last week's report. And in your inbox are requests that you have to answer all just a little differently. That is usually not difficult work. It does return every week, and it costs you exactly the hours you wanted to give to your clients.

You might recognise this. I see it with many freelancers I meet. They are good at their profession, and yet they are stuck in the evening with the manual work around it. You are probably already wondering if that can ever be different.

Your advantage lies in working more personally. AI helps you work faster and more tailored to your client, with your own voice included.

After a training session, all participants usually get the same email. Thank you for attending, here are the slides, see you next time. That is polite and everyone does it that way. It is almost funny: the more polite and general the email, the faster you forget it.

Think about an email that is different for every participant. One that refers to what that person actually said during the day. And one that asks what they can move forward with now. In terms of content, that is a small step. In the past, you simply could not take that step, because you did not have the time for it.

Many small costs keep running because no one looks at them. Which tasks you can take into your own hands with AI, and how you start small without stacking subscriptions.

At the end of the month, you sit with your bank statements. There is a subscription, an invoice from the freelancer and the virtual assistant. Next to that, you see the little tool that types out your conversations and another tool to schedule your social media. Each amount on its own is not too bad. Together they form a row you would rather not look at for too long.

Ask yourself what you are actually paying for. Sometimes it is a profession that someone else can do better. Sometimes you just never got around to it yourself. You probably recognise this, and perhaps you are already wondering which of those expenses belong to the second kind.

Two colons and a short word, and your AI knows what you mean. This is how I put custom commands in my custom instructions, for things I ask every week.

In a workshop this week I showed my custom instructions on the screen. It says that I do not want to see any emojis. And not too many bullet points, and normal capital letters. And below that is a list of lines that start with two colons. Those are the rules I use most often, and they are barely a word long.

Feel free to copy the prompt at the bottom of this article, and you can write your first shortcut together with your AI.

What you type again every week

Take a look at your own chat history. Chances are you formulate the same questions over and over again. Translate this to Dutch. Write a short bio about me. Make this text a bit looser. Every time you type a sentence, sometimes two. And every time it is slightly different, so the result is too.

You do not write the best prompt alone. With one sentence at the end, you let AI question you first, and you become sharper in your thinking.

People often ask me about the perfect prompt: which template, which order, which magic words. You might have wondered about that too. My answer surprises most participants, because I do not write my best prompts alone. I let AI question me first.

Copy the prompt at the bottom of this article. Further down is also the prompt package with six metaprompts that question you first, free for Members.

One sentence at the end of your prompt

The technique is called the metaprompt, and it is actually almost embarrassingly simple. You describe what you want to achieve and what a good result looks like. Then you add one sentence: 'Ask me three to five questions first before you start.'

A long chat does not train your AI, it fills the context window. Why the quality drops halfway through, and the habit that solves it.

Perhaps you have been working in the same chat for weeks. In almost every workshop someone tells me that, and usually with some pride: 'That is how it gets to know me.' I understand that thought, because it feels like a colleague who knows better and better how you work. Only it is not true, and precisely because of that the answer gets worse instead of better after a while. That is a bit ironic, because you are doing it exactly to make it better.

Copy the prompt at the bottom of this article to neatly wrap up a long chat and start fresh.

A simple peer coaching exercise using the colours of the Six Thinking Hats. You just hold up a card and learn to listen in the meantime.

Perhaps you want to help a colleague with a difficult decision and you quickly offer solutions. We all do that. There is one exercise from my creativity workshop that participants almost always take home with them. You need six coloured cards for it and one strict rule: you must not give advice.

Copy the prompt at the bottom of this article if you have no one to practise with, and let your AI hold up the cards.

How the exercise works

You work in a duo or a trio. One person talks about an idea or a challenge they are working on. That must be something real, not a fictional case. The other listens, with six coloured cards in their hand.

Four habits that turn an AI strategy into daily practice, for individuals, teams and entire organisations.

You have an AI strategy, or your team has met about it at least a few times. And yet not much is changing in your working week. You probably recognise this. EPIC helps you turn that strategy into daily practice, for yourself, for your team or for your entire organisation.

Copy the prompt at the bottom of this article and apply the four habits to one task straight away.

Why strategy alone is not enough

Most AI adoption processes fail for the same reason as most change processes. They focus on the tools and forget the people. That is a bit ironic, because the tools are there precisely for those people. EPIC starts somewhere else: with the work you are already doing. That way you do not have to start a separate project. It is about four habits: Everyday tasks, Pair learning, Iterative feedback and Continuous improvement.

Five human capabilities that AI cannot take over, and a prompt to help you discover which ones you already use in your work.

You know those workdays when nothing goes as planned. The situation is messy, the direction changes, and nobody knows exactly what is expected of you now. That is when you need cognitive agility. It is the human ability to adjust your thinking and your emotional and social approach. This is how you find your way in unpredictable situations. It is a meta-capability that consists of five capabilities.

Copy the prompt at the bottom of this article to see with your AI which of those five capabilities you already use.

Five capabilities, one meta-capability

Seven questions you answer before you press enter: this is how you turn a vague prompt into a reliable brief, for older and simpler AI models.

You type a short question into your AI and get something back that looks neat. Only it is not what you meant. So you try again, and again, until you refine the result yourself. You probably recognise this. The CRAFTER framework is a fixed way to brief your AI, and it ensures this happens less often.

With a loose prompt you sometimes just get lucky. With CRAFTER you actually get the same quality every time. You turn a vague prompt into a clear task, and you can repeat it.

A side note beforehand: CRAFTER is a valid framework for older and simpler AI models. For ChatGPT 5.x and Claude 5.x it no longer applies. That is why you will also find a prompt for the more recent models at the bottom.

From Status Quo to Elevate: five levels that show how professionals and teams grow their relationship with AI, and what each level requires of people.

You might have been using AI for a while, and you wonder where you actually stand. Or where your team stands. The Superworker Model maps out five levels of how professionals and teams grow their relationship with AI. Speed or efficiency say little about that. The model looks at what becomes possible at each level, and at what each level requires of the people in the system.

Copy the prompt at the bottom of this article and put your own work on the map, task by task.

A map of possibilities

The five levels are called Status Quo, Optimize, Redesign, Reinvent and Elevate. I see the model as a map of possibilities. Different parts of your work can be at different levels. So you do not have to reach the highest level everywhere. I would consciously choose where you put your energy and attention.

Why your judgement gets weaker if you give all your thinking to AI, and why human attention remains the scarce computing power.

You let AI do more and more of your thinking, and that goes smoothly. What happens to your own thinking in the meantime gets too little attention in the AI conversation. Perhaps you do not notice anything yourself yet.

Copy the prompt at the bottom of this article. It helps you choose which thinking you keep for yourself.

What is The Drift?

Perhaps you outsource cognitive work to AI, such as research and analysis, or writing and decision support. Then something subtle begins to happen. The muscles you do not use begin to waste away. That actually happens very steadily. There is no day on which you feel the difference.

A chatbot waits for your question, an agent completes a goal. Here is how you build your first own AI workflow this week, with a prompt to help you.

You probably already use AI to ask questions, and perhaps you wonder what comes next. Meanwhile, a shift is happening. Many people have not noticed it yet. We are moving from asking AI questions to directing AI workflows.

Copy the prompt at the bottom of this article, and you will map out your first workflow together with AI.

What is the difference between AI agents and chatbots?

A chatbot waits for your input and gives you an answer. An AI agent gets a goal, breaks it down into steps, uses tools and delivers a result. It is a bit like the difference between asking someone for directions and hiring a driver who knows the route.