Skip to content
← Journal — Project

AI training: learning by doing, with Ariane

In many AI training courses, people watch more than they practise. I created Ariane, an assistant that supports every participant, attempt after attempt.

By Thierry Lechanteur · · 2 min read

We watch a lot. We practise too little. That is the trap of many AI training courses, mine included. To address it, I created Ariane, a training assistant that supports every participant during the workshop: she analyses their prompts, spots what isn’t working and helps improve them, explaining why.

The trap: watching without practising

I have been teaching AI since 2023. I collected certifications, but on the teaching side, I was left wanting more. In a room, the trainer shows and the participants watch. Everyone leaves with notes, but too few attempts of their own.

Yet you don’t learn to work with AI by watching someone else do it. You learn by trying, getting it wrong, and understanding why one result is better than another. I was missing a tool to give more room to that practice.

An assistant for every participant

Ariane supports each person, attempt after attempt. When a participant writes a prompt, she analyses it, spots what isn’t working and suggests an improvement, explaining her reasoning.

While I lead the group, everyone gets personalised support to experiment and progress at their own pace. Nobody waits their turn, and nobody is left behind.

Ariane training poster: “We watch a lot. We practise too little.” Personalised support to learn by doing. Steer, phrase, check, own.
Ariane, my training assistant: the common thread fits in four verbs.

A prompt library built together

The most interesting prompts are shared with the whole group. At the end of the course, everyone leaves with a library built together, fed by everyone’s attempts and discoveries. It keeps serving day to day, long after the workshop.

The common thread: steer, phrase, check, own

Learning to work with AI means learning to stay in control. The whole course follows four verbs:

  • Steer: decide where AI comes in, and stay in charge.
  • Phrase: know what you are asking it, clearly.
  • Check: judge what it suggests, with a critical eye.
  • Own: decide what you do with it, and stay accountable.

A course, a workshop, a lab, together

Ariane brings together a course, a workshop, a lab, a library and a space for dialogue in a single interface. The goal: learning AI in a practical and playful way, without spending the day watching a projected screen.

For your team

I offer two formats:

  • Training your team in AI with Ariane and me, on-site in Liège, in Brussels or elsewhere in Belgium, or remotely.
  • Using Ariane for your own training, if you train your teams or clients yourself.

Discover our AI workshops and training, or write to me to talk about it.

Give us 30 minutes,
we handle the rest

Free brief, no commitment. We will tell you honestly if we are the right studio for your project — and how much it costs.