This page was translated from Dutch with AI.
Model, app, environment: know which layer you are in
At the end of a long workshop day, I asked the group what they were taking away. One participant said something I have not forgotten since. He had used two programmes side by side all day, and only now realised they were two different things. Another participant did not see the added value of an agent app, because 'I already have ChatGPT'.
They are not stupid questions. They are the questions everyone has, only not everyone dares to ask them. And they all have the same cause: we talk about AI tools as if they are one thing, while there are actually three, neatly slid into each other.
The three layers
I always draw it as three circles around each other.
The inner circle is the language model. That is the brain that predicts the text: GPT, Claude, Gemini and their family. You never see it directly. You always talk to it through something else.
Around that is the app. That is the programme in which you use that model. The chat windows everyone knows, but when you start building also code editors like VS Code or Cursor, or an agent app like Craft Agents. The app determines what the model can touch: only your chat, or also the files on your computer.
And around that again is the environment you build. That is often just a folder on your computer, with your projects, your instructions, your skills and your agents in it. That is your armour around the model, and it is the layer that makes the most difference.
Keep reading for free
Register with your e-mail address to read the whole article. It is free, and it opens every article and programme in the learning space.
Check your inbox
We've sent you a sign-in link. It works for 30 minutes. Nothing there? Check your spam folder too.