VITON13School
AI course

Learn AI by building: what 13 missions produce on your own computer

Learning AI by building means every lesson ends with something that works on your own computer, such as a prompt library, a private chat or a research agent. You practise on real tasks, compare weak and strong prompts, check yourself with quiz questions and keep the files, so the skill stays after the lesson ends.

By , Founder of VITON13 SCHOOL
Six live scenes from the AI course page: writing, drawing, code, voice, an agent and answers from files

Watching someone use AI feels like learning. Then you open a blank chat and nothing comes out. The gap is practice with your own material. This guide explains why building works, what the 13 missions of the course produce and how the free start and code access work.

Key ideas

  • Every mission ends with a working result on your own computer, not a certificate alone.
  • The lessons contain 40 prompts, 13 hands-on tasks and about 7 hours of practice.
  • Missions 1 and 2 are free; the other eleven open with a personal code.
  • It is for people who want to use AI at work, not to collect theory.

Why building beats watching

A video shows a finished result and hides the dead ends. When you build, you meet the dead ends yourself: the model that invents a source, the prompt that works once and fails twice, the file that will not open. Solving those is the actual skill.

Building also leaves something behind. A saved passport, a prompt library or a working chat is useful next week, when a certificate would not be.

[1]

What the 13 missions produce

Each mission has a named build. Mission 1 gives you Your AI Passport, and mission 2 Your model stack. Mission 3 is Your prompt library v1, and mission 4 a verified research memo. Mission 6 is a landing page you can open and ship, mission 7 a private chat running on your machine and mission 8 your document chat.

Later missions build an inbox-to-Telegram digest, a competitor research agent, a voice-note-to-tasks pipeline and your bot’s safety test pack. Mission 13 ends with Your AI operating panel, which pulls the pieces into a personal system.

The practice, counted

Counted from the lessons themselves, the course holds 40 prompts to copy, 13 hands-on tasks, about 7 hours of practice, 13 weak-versus-strong prompt duels and 39 quiz questions. It also points to 42 open-source GitHub projects, with star counts read on 16 November 2026.

The duels matter most. You see a weak prompt and a strong one side by side and learn why the second works, which is the habit you carry to any model.

Free start, code access and who it is for

The course opened on 17 November 2026 in four languages: English, Russian, French and Spanish. Missions 1 and 2 are free. The other eleven open with a personal code on your VITON ID account, and a code works for one account. The whole course costs 1 300 ₽.

It suits people who use AI at work or for study and want repeatable results: writers, analysts, students, small-business owners, and anyone curious enough to build. No income or job is promised; the promise is smaller and checkable, a working result per mission. Start with mission 1 and see for yourself.

[2][3]

In everyday language

Learning to cook from a video is like watching someone else eat. You learn when you hold the knife, burn the first pan and fix the second. Each mission puts the knife in your hand, with a recipe you can read and a dish at the end.

Try it yourself

Open mission 1, run the passport interview and save the result as a file on your computer.

Expected result

A saved passport file that two different chat apps follow, which proves the lesson worked.

Check your answer: What does every mission of the course end with?

A result that works on your own computer, such as a passport, a private chat or a research agent.

Questions

Do I need to pay before I try the course?

No. Missions 1 and 2 are free. The other eleven open with a personal code on a VITON ID account, and the whole course costs 1 300 ₽.

Do I need programming experience?

The missions are written as steps you can follow and each ends with a working result, so you do not start from theory. Later builds are harder, and you move at your own pace.

Start with mission 1, open free on VITON13 SCHOOL, and keep what you build. If it works for you, the other eleven missions open with a personal code for 1 300 ₽ for the whole course.

Sources and further reading

  1. VITON13 SCHOOL — Course 01: Artificial Intelligence ↗Sources checked:
  2. Anthropic — Prompt engineering overview ↗Sources checked:
  3. Ollama on GitHub ↗Sources checked:
VITON13 SCHOOL / AI course

Artificial Intelligence

The broadest and the deepest programme at once: how to work with artificial intelligence deliberately — from framing the task, through prompting and verification, to automating your own processes, building agents and assembling a system that is yours.

Explore the programme ↗