Why the subject matters
Someone reading CVs sees a dozen identical online shops a week. Not because they're bad, but because everyone builds them from the same handbook. A subject of your own gives you an edge before they've opened the code.
Finding your own
The most reliable method is to remember what personally annoys you. What do you do by hand every week? What do you forget? What do you dig out of old messages because there's nowhere else to look?
The best projects grow from a personal irritation: they have a real user — you — and you can tell immediately whether the solution works.
Twenty subjects
For yourself:
- Expense tracking from photos of receipts.
- A reminder for renewing subscriptions and insurance.
- Search across your own notes and bookmarks.
- A reading tracker with a readable export.
- A household chore calendar split between family members.
For a small organisation:
- An intake form for a shelter or charity.
- A timetable with booking for a small studio.
- Inventory tracking for a club.
- A catalogue for a one-person craft business.
- Collecting and processing entries for a local event.
About data:
- Tracking prices of something specific and charting the change.
- Turning your own purchase history into readable reports.
- A map of something civic: drinking fountains, playgrounds, benches.
- Analysis of your city's open data.
- Comparing published transport timetables with reality.
Tools:
- A bot that sends a timetable or reminders.
- A utility for bulk renaming and sorting files.
- An uptime check for a site, with alerts.
- A converter you personally keep missing at work.
- A small dashboard of your own numbers: sleep, exercise, spending.
Subjects to avoid
Not because they're bad, but because there are too many of them: to-do lists, weather, blogs, an online shop with no idea behind it, another chat app.
If you really want a well-worn subject, add an unusual constraint. A to-do list that orders tasks by deadline and your actual workload is a different conversation.
Choosing from the list
Pick something you'll use yourself for at least a month. That's the only reliable signal: a project you need gets finished, while one taken "for the portfolio" gets abandoned in week three.
And judge the size honestly: if the idea doesn't fit into six or eight weeks, halve it. A finished small project always beats an abandoned large one.
Хватит читать — пора делать
На CohortX можно найти команду под пет-проект и получить тот самый опыт, о котором спрашивают на собеседовании.
Похожие статьи
- Presenting a repository so people actually look at itNaming, structure, commit history, what to hide and what to pin — the small things that decide whether your code gets opened.
- Как оформить репозиторий, чтобы его смотрелиНазвание, структура, история изменений, что скрыть и что закрепить — мелочи, которые решают, откроют ваш код или нет.
- A README that makes people open the projectWhat to put in your repository's README, in what order, and why without it nobody looks at the code.