Is this training or coaching?

Both, depending on who is learning. A team works in collective sessions on its own cases. A single person moves forward in recurring sessions, on an objective set at the start.

My team is not technical. Does this work?

Yes. We start from the tasks that come back every week, not from models. Nobody needs to code to pick the right tool, give it the right context and check what comes out.

We already know how to use ChatGPT. What else is there to learn?

How to leave the chat window. Working on your real documents, connecting a model to your tools, having a sequence of actions executed, and knowing when a human has to take over.

Which tools are covered?

The ones worth your time at the time of the session, not a frozen catalogue. Today that means the main models, agentic environments and n8n for orchestration. The landscape moves fast, the programme follows.

How long does one-to-one coaching last?

As long as the objective keeps moving. We set the target in the first session and hold the same pace until it is reached. There is no open-ended commitment.

What if we find out a system needs to be built?

It happens, and it is not pushed. We can stay on the levelling-up work, or scope an agent and workflows separately if the need is clear.

The AI training that moves you from prompting to agents.

Almost everyone can ask a model a question. Far fewer people know how to pick the right tools, give them the right context and put agents to work. That gap is what we close, with your teams or with you alone.

Two ways to close that gap.

Your teams

Move the whole team to advanced use.

Everyone improvises with the tool they happen to know, nobody shares a method, and the capabilities that really change the work stay unused.

What we work on

The tools that earn their place, working on your documents and your data, first agents on the tasks that used to come back every week, and what has to stay validated by a human.

How it runs

Small-group sessions, on your own cases, with material the team keeps and reuses.

You, one to one

Go as far as your objective requires.

You already know how to use a model. What is missing is the next step: agents, your own stack, or a business to build around AI.

What we work on

The objective is set in the first session, then we work on it: build your stack, move to agents, industrialise an internal use case, target a vertical.

How it runs

Recurring sessions, at the pace the objective requires, until it is reached.

We do not teach you AI. We teach you to put it to work.

Skill is not measured by how many tools someone knows, but by what they can keep running without going back to it. So we start from what you already do, then add what is missing for the work to come out without you.

Pick the tool for the task, not out of habit.
Give a model your context: documents, data, business rules.
Move from an answer to an execution, with human validation in the right place.

Who this is for.

The point is not company size, it is the gap between what you do with AI today and what you want to be able to do with it.

Business leader

You want the team to stop experimenting on its own and work with a shared method.

Executive or manager in a large group

You run a division or a business unit in a CAC 40 group and you want a regular counterpart, not an annual seminar.

Investor or fund

You want to judge an AI deal on what holds up technically, not on the promise in a deck.

Developer

You want to build with AI and agents, not use it as a slightly longer autocomplete.

Independent

You want to build an offer around AI, pick a vertical and know what you are selling.

Employee

You want to get ahead in your own job without waiting for your company to move.

Pace and format are decided during scoping: it is not the same exercise for a team of ten and for a single person.

If you would rather we handled it.

AI agents and automation

When you want the system in production rather than learning to build it yourself.

See the agents offer

Illustration of the AI audit and consulting offer

AI audits and consulting

When the question is not how to do it, but what deserves doing in your company.

See audits and consulting

Or try an agent on your site

What people ask before starting.

One call, one objective.

Come with what you want to be able to do in three months. We look at whether that is a team session, one-to-one follow-up, or neither.

See how Kirako runs day to day