Building AI Agents

Enough to have called an LLM API and seen what comes back. The course does not spend time on what a model is; it starts where you want one to do something on its own. Strong general engineering experience matters considerably more here than AI experience does.

We build with one and also without one, because the point is understanding the loop rather than a particular library's opinion of it. Tell me what your team already uses and the exercises run there. If nothing has been chosen yet, this course is a good place to make that decision with evidence behind it.

Yes, and the four-day version is largely that. Bring a real workflow you are considering handing over. Some groups conclude during the course that theirs should stay a script, which is a genuinely good outcome and far cheaper to reach in a workshop than in a quarter of engineering time.

Modest, but not nothing, and agent runs cost more than chat does. I send an estimate along with the requirements once we have a date. If you would rather cap it, most of the material runs against smaller models without losing the point.

Five to twenty. Agent work is slow to run and quick to argue about, so the group has to be small enough that everybody's build gets looked at properly at least once.

Yes, in German or English. Materials stay in English because the libraries, their documentation and every error message participants will meet are in English, and a translated glossary tends to make those harder to search rather than easier.