Skip to main content
✍️

Python practice — not a video course you'll never finish

Video courses sell well and stick badly. Active recall — code you type, errors you debug, predictions you check — is what actually moves the needle.

The dropout numbers for video-only programming courses are brutal: 5-10% completion on average across the major platforms, much lower for anyone who's not already a developer. The problem isn't the content quality — it's the medium. Passive watching activates recognition, not recall. A week later you remember the instructor's voice and almost nothing about decorators.

Every lesson here makes you do one of four things: write code that has to pass a test, predict the output of code you're shown, find a bug in working code, or complete a fill-in-the-blank. The AI tutor only opens its mouth when you ask, and even then leads with a guiding question. First 15 lessons are wide open — no signup, no card — so you can test the format before committing.

Start lesson 1 — write code, don't watch a video

Or start with one of these

7-day jumpstart
35 hands-on exercises, day-by-day.
Platform stats
1224 lessons, 18 languages, real counts from the codebase.