How it's made

Gen2Courses courses are generated with AI, from the outline to the narration. What makes them trustworthy is that the pictures come from running real code.

One visual language

Every data structure has one shape and one colour, in every episode of every course: an array is a row of touching blue cells, a String is the same row in cyan with quotes, a HashMap is an orange panel of key → value rows, a stack is a cup open at the top. Pointers, windows and highlight colours always mean the same thing. Learn it once, in the first episode of a course, and every later picture reads at a glance.

The animation is the algorithm

Each solution is written as ordinary code that records a snapshot at every step: which line ran, what changed, and what to say about it. The video plays those snapshots back. Because the animation is produced by the algorithm itself, the numbers on screen, the highlighted line of code and the caption always agree.

Checked against real systems

The database course goes further: every claim about Postgres, from how many pages a B+ tree reads to what a lock waits for, is shown with output captured from a real Postgres server running the same queries.

Made with AI

The courses are planned, scripted, animated and narrated with AI tools (Claude writes the lessons and the animation code; Remotion renders the video; a neural voice reads the captions). Spotted a mistake? Leave a comment on the video on YouTube.