The one signal that matters
Whether live humans review your code and give substantive comments.
Everything else — the syllabus, the platform, the length, the brand — is secondary. A course without work being reviewed is a set of videos, and identical videos exist for free.
Ask directly: who reviews, how many pieces of work, in what form, and how long you wait for a reply.
What else to check before paying
Who teaches. Practising professionals, or instructional designers retelling a textbook. Look at where these people work now.
How much practice. The ratio of watching to building. Twenty hours of video and three exercises is a lecture series.
The final project. If every graduate's is identical, it's useless in a portfolio. Good courses let you choose the subject.
Whether there's team work. Rare and very valuable: it's exactly what course graduates lack.
What happens afterwards. Do you keep access to materials, can you still ask questions.
Promises that should worry you
- "Guaranteed employment". Almost always means help with a CV and a mailing list. Read what's actually guaranteed and what counts as fulfilment.
- "Zero to hired in three months". Such timelines belong to a handful of people.
- "Money back if you don't find work". Usually hedged with conditions: so many applications, so many assignments, a deadline for claiming.
- Pressure and discounts expiring today. A decent school doesn't rush you.
- Reviews only on their own site. Look for independent ones.
How to test before paying
Take the free portion. Almost everyone has an introductory module. It shows the teaching style and the quality of materials.
Find graduates. Not the ones quoted on the site — people in professional communities. Ask directly what helped and what didn't.
Look at graduates' projects. If they're all the same, yours will be too.
Ask about code review. The answer to that question usually settles everything.
What it costs and when it's worth it
Courses cost the equivalent of several months of living expenses. Paying makes sense if you need structure, discipline and genuine code review.
It doesn't if you can learn on your own and are simply looking for the "correct route" — the route is publicly available, and the money saved buys books and time.
What to do alongside a course
From month one, build your own things, off-syllabus. Otherwise you'll finish with a certificate, the same final project as everyone else, and a complete inability to start from a blank file.
And find somewhere your code will be read by people other than your assessor: open source projects, a team of like-minded people, a shared project. It's free and often more useful than the course itself.
Хватит читать — пора делать
На CohortX можно найти команду под пет-проект и получить тот самый опыт, о котором спрашивают на собеседовании.
Похожие статьи
- Python from scratch: what to learn and in what orderThe order of topics from variables to a first service, what can wait, and where people usually get stuck.
- Python с нуля: что учить и в каком порядкеПорядок тем от переменных до первого сервиса, что можно пропустить на старте и на чём обычно застревают.
- How to ask questions that get answeredThe shape of a good question, how long to struggle alone, and why asking well is valued as a skill in itself.