We've opened an evening practical AI school

We've opened enrollment for our evening practical AI school - a program for individuals who want to stop guessing with AI tools and start using them safely, concretely, and every day. AI literacy is no longer a bonus on your CV; it's becoming a basic working competency. This school exists so you can build it on real tasks from your own job.

The school is open - what that means

We deliberately kept the format compact and practical:

• Online and in the evening, four sessions a week (Monday to Thursday) across three intensive weeks - twelve sessions in total.
• Groups of up to ten participants, so everyone gets feedback on their own work.
• Eight modules - from the basics of AI and prompting, through generative tools and agentic processes, to your personal implementation plan.
• Six starts a year, a new cohort every two months. Classes are in Croatian; English cohorts are arranged on request.

Each session is built around exercises with your real tools and tasks - what you learn on Monday, you apply on Tuesday.

Why AI doesn't make you productive on its own

The most common misconception is that AI saves time the moment you open it. It doesn't. A weak prompt returns a generic answer that you then spend longer fixing than if you'd done the task yourself. The difference between "AI wastes my time" and "AI doubled my capacity" is almost never the tool - it's how precisely you know how to ask.

Technology is here to empower people, not replace them. Productivity only arrives once you learn to lead the tool rather than wait to see what it offers.

Anatomy of a good prompt

A quality prompt almost always contains four things: context, a role, a clear goal, and the desired format. Compare:

• Weak: "Write me an email to a client."
• Good: "You are a project manager. Write a short, warm email to a client (max 120 words) letting them know we're two days late due to extra testing, apologizing, and proposing a new deadline on Friday. Tone: professional, no excuses."

The second prompt doesn't ask the model for anything smarter - it just gives it everything a person would need anyway to do the task well.

Five habits of productive work with AI

1. Give context before the task. Who you are, who you're working for, what the goal is, and what you're worried about. The model doesn't read minds - it reads what you write.
2. Ask for the format explicitly. A table, a list, a five-point outline, JSON - say what the result should look like so you don't have to reshape it afterward.
3. Work iteratively. Ask for a first version, then steer with successive directives ("cut it by a third", "add an example", "change the tone"). A chain of steps beats one perfect prompt.
4. Always check the result. AI sounds confident even when it's wrong. Verify facts, figures, and names before you send anything out.
5. Build your own prompt library. Save the prompts that once produced a good result and reuse them. That's how personal productivity turns into a repeatable standard.

What you'll learn

Across eight modules we take you from the basics to your own implementation plan. You'll learn:

• how to use AI to boost productivity in everyday work;
• how to phrase prompts and tasks properly and structure templates you can reuse;
• how to judge the quality of an AI answer and recognize hallucinations;
• how to protect business and personal data;
• how to bring AI tools into your own role through a concrete 30-60-90 plan.

The methodology is the same as in our corporate program - only the format changes, adapted to an individual's evening rhythm.

Enrollment is open

Places are limited to ten participants per cohort, so we fill groups in order of application.

See the program and dates on the evening school for individuals page, and when you're ready, apply directly via the contact form. Once we receive your application, we send a short questionnaire to map your existing knowledge and the work you want to bring AI into.