CohortX
Блог

A two-week pet project: realistic bounds

2 августа 2026 г. · 2 мин чтения · Читать по-русски · Антон Молотило

Why two weeks is a good length

Interest doesn't have time to cool. The dip arrives around week three. If the project is done by then, you didn't abandon it.

There's a visible result. Two weeks in you have a working thing, not "started in May, sign-up is done".

It repeats. Three short projects in a quarter give you three finished pieces and three things to talk about. One long one usually gives nothing.

What you can finish

One path end to end. For example:

  • a bot that takes a message and files it into a spreadsheet, with one meaningful command;
  • a page listing something, with a form to add items and storage behind it;
  • a utility that parses files and produces a readable report;
  • pulling data from one source and charting it simply.

What they share: one user action, carried through to the end, deployment included.

What you can't

Sign-up and login with password recovery, roles and permissions, payments, notifications, a mobile app, a pixel-perfect design.

Not because those are hard. Because each adds a week, and together they don't fit into two.

Choosing the idea

Take your idea and cross out everything except one action. It usually hurts, but the crossing out is what turns "I want to build a service" into something finishable.

A check: if explaining the idea takes more than two sentences, cut further.

A two-week plan

Days 1–2. Agree what you're building and what counts as done. Set up the environment, create the repository.

Days 3–4. Deploy an empty skeleton so it opens behind a link. Early, deliberately: deployment left to the end always costs more than expected.

Days 5–10. The main scenario. Only that, nothing extra.

Days 11–12. Tidying: errors, empty states, comprehensible messages.

Days 13–14. README, a short write-up, and a check that everything opens from someone else's device.

Why it goes better in a team

Two weeks is the span where agreements still hold and fatigue hasn't accumulated. People rarely vanish over such a stretch.

Short projects have another pleasant property: everyone's contribution stays visible. Over two weeks it's only unclear who did what if you worked in silence. With normal check-ins it's easy to describe at the end who owned what — and that's already a record of experience, not a vague "we built it together".

Хватит читать — пора делать

На CohortX можно найти команду под пет-проект и получить тот самый опыт, о котором спрашивают на собеседовании.

Похожие статьи