What's no longer worth learning
Courses, prompt packs and middleman tools: what this series of articles has been throwing out.
7 cases59 min of readingLast case: 25 Sep 2026
If you've just arrived in the world of AI, chances are your head is full of videos promising that everything is dead easy. Here we start with the basics, but told from real use: which AIs are out there, how they differ, which one is worth paying for and which one you don't need. I work mainly with Claude, but also with OpenAI, DeepSeek, GLM, Kimi and MiniMax, and in this section I explain what I use each one for and why I don't stick to just one.
What gets something into this section isn't that it's simple, but that it helps you build judgement and a basic way of working: choosing a tool, asking well and knowing when to trust the answer. So the articles about a specific AI are here to help you decide, not as manuals.
We also talk about money: can you work with AI without paying? Up to a point. I'll show you where the limit of each free version is, what gets cut off when you hit it, and what you pay without realising, which is often your data or your time. And when I do pay, I tell you how much and why.
And how to ask AI for things: I've tried the "hidden tricks 90% of people ignore" and almost all of it is hot air; what's left is a handful of principles that really do change the result. Starting from zero, a good order: "Are chats still useful?", then "Asking AI for things without magic formulas", then "Why I switched from ChatGPT to Claude" to choose a tool and, finally, "Can you work professionally with free AI tools?".
Courses, prompt packs and middleman tools: what this series of articles has been throwing out.
I tried the hidden tricks and the perfect prompt. What's left: a few principles that really change the result.
With agents and harnesses around, why I still open a chat window.
What they do well, what they cost, and what data I give them and what I don't.
ChatGPT, Claude, Gemini, Copilot, DeepSeek, GLM, Kimi and MiniMax, and where the limit lies.
What made me switch, what I miss and what I still use OpenAI for.
The setup behind everything I write about on this blog.
One email a week with what was verified, what was refuted and what is still on the bench. No news roundups.