Name: MARS IT School
Address: Tashkent, Tashkent, UZ
Telephone: +998 78 777 77 57
Price range: $$

MARS IT School methodology — the lesson standardA MARS IT School lesson lasts 60 minutes and has seven blocks: hook, recall, one new idea, practice, quiz, second practice and a show of results. Children work on their own for at least 80% of the time, and every lesson ends with a working result. AI is used only under the rule “explain every line”. Homework continues the lesson’s project and is checked before the next lesson.Every lesson ends with something to showThis is the open standard every Mars branch teaches by. It sets out how a lesson is built, what a child does in it, and what we deliberately gave up.Five principles

Children spend eighty percent of the lesson doing. Explanation takes no more than a fifth of the time. A lesson plan with a longer lecture is sent back for rework. Skill comes from hands-on work, not from listening.

Every lesson ends with a working result. A website, a game, a character, a bot — a small but real product that works. Once a month a child publishes a piece of work with a link they can send to parents and friends. “We covered the topic” is not a result. A result can be opened and shown.

The child is the author; AI is a tool. We bring in AI and every other tool with one rule: explain every line it wrote. Mastery means “I understand it and can change it”, not “it worked”. Work a machine did without understanding creates a false sense of skill and kills the wish to learn the basics.

The group moves together; pair work is a normal part of the lesson. A strong student works with a beginner; children review each other’s work and add features to a neighbour’s project. Explaining to someone else is the best way to understand it yourself.

Real tools at a child’s level. As they grow, students work with what the industry uses: Git, Figma, task trackers, AI assistants. Every lesson answers “why”, not only “how”.

The lesson, minute by minuteA lesson lasts 60 minutes. The order and set of blocks are the same in every branch; the mentor adjusts timing within blocks to the group.

Hook (2–3 min): A live demo of the lesson’s result, an unexpected question or a “broken” example. A lesson never opens with “today’s topic is…”.

Recall (4–5 min): 3–5 questions on past lessons. Children recall on their own, with no answer slides.

New idea (7–8 min): One new idea per lesson. The mentor shows it live and thinks aloud: “I’m writing this because…”

Practice I (18–22 min): Worked example → example with gaps → their own version. The mentor helps with a question, not a ready answer.

Quiz (3–5 min): 2–4 questions that catch typical misunderstandings. The answers tell the mentor whether to move on or unpack the mistake.

Practice II (13–17 min): Their own feature in the project: pair work or an AI step where every line must be explained. A harder task is ready in advance for fast finishers.

Show (4–5 min): The work is shown to a neighbour or the group. Homework comes with clear completion criteria.

Recall and the quiz are never cut, even when time is short. They are the cheapest and best-proven techniques in teaching.Two ages, two rhythms

Younger groups (7–11): continuous work — Up to 15 minutes, then a change of activity; explanation in one go — Up to 5–7 minutes; instructions — One step at a time: show, then do; practice — More templates and visual metaphors; pairs — Set up and guided by the mentor

Older groups (12–17): continuous work — Up to 25–30 minutes; explanation in one go — Up to 12–15 minutes; instructions — Two or three steps ahead is fine; practice — More open tasks and choice; pairs — Rotate; children organise themselves

HomeworkHomework always continues the lesson’s project rather than setting a separate abstract task. Each assignment has 3–5 checkable completion criteria, so the child knows when they are done. It takes 20–40 minutes, is handed in through the SPACE platform, where parents see it too. The mentor checks it before the next lesson.Every lesson is scripted in advanceMentors don’t improvise. Every lesson of every course has a script written by a methodologist:

A minute-by-minute plan following the blocks above.

Exact wording for key explanations: what to say and what to show on screen.

Typical mistakes children make and how to respond to each one.

Project code in three states: at the start, after the mentor’s demo and at the end.

A task for fast finishers and a simpler path for those who struggle.

That is why a lesson in Yunusabad and one in Chilanzar follow the same standard, and quality doesn’t depend on which mentor a group gets.What we gave up

Lectures longer than a fifth of the lesson. And “read this and answer the questions” tasks.

Retyping code from the screen. The child always adds their own solution.

Lessons without a result. A lesson with nothing to show counts as a defect.

Unstructured self-study. Most children drop such courses.

Points instead of interest. XP and levels in SPACE show how the child is growing: what they have made and learned. It is a map of their progress, not a prize for a grade.

“Look, the AI did it all by itself.” Demoing AI without understanding every line is not allowed.

What it rests onEach block of the lesson rests on a technique with proven effect. The main sources:

Recalling without prompts consolidates material better than rereading. Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning. Psychological Science, 17(3).

Worked examples with gradually faded steps work better for novices than solving from scratch. Renkl, A., & Atkinson, R. K. (2003). Structuring the transition from example study to problem solving in cognitive skill acquisition: A cognitive load perspective. Educational Psychologist, 38(1).

Active learning produces markedly better results than lecturing. Freeman, S., et al. (2014). Active learning increases student performance in science, engineering, and mathematics. PNAS, 111(23).

A short check of understanding during the lesson lets mistakes be fixed in time. Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education, 5(1).

External rewards for an interesting activity reduce intrinsic motivation. Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125(6).

Only a small share of those who enrol in open online courses finish them. Reich, J., & Ruipérez-Valiente, J. A. (2019). The MOOC pivot. Science, 363(6423).

Courses · Contacts · Book a trial lessonPhone: +998 78 777 77 57

MARS IT School — Toshkentdagi IT o‘quv markazi va dasturlash maktabi MARS IT School 7–17 yoshli bolalar va o‘smirlarga dasturlashni o‘rgatadi: veb-dasturlash, Python, o‘yin yaratish va robototexnika. Darslar Toshkent filiallarida oflayn, 12 tagacha o‘quvchidan iborat guruhlarda o‘tadi. Birinchi sinov darsi — bepul.

Kurslar

IT Kids — 7–12 yosh

IT-dasturlash — 12–17 yosh

Maktab haqida

Kontaktlar va filiallar

Sinov darsiga yozilish

Telefon: +998 78 777 77 57