A list, not a course, and the most famous one there is
First, set the expectation: Build Your Own X is not software and not a course. It is a curated index, an awesome-list, of tutorials that teach a technology by having you recreate it from scratch: build your own database, git, Docker, regex engine, operating system, programming language, and dozens more. At well over half a million stars it is one of the most-starred repositories on GitHub, and that popularity is itself the signal. The “rebuild it to understand it” approach is the single most reliable way to turn surface familiarity with a tool into real understanding, and this is the canonical map of where to do it.
The pedagogy is the reason it endures. Reading about how Docker works leaves you able to nod along; writing a tiny container runtime yourself leaves you able to reason about it. This repository does not teach that itself, it points you at the people who do, organized by the technology you want to demystify.
What is actually in it
The breadth is the draw. The list is organized by the technology you rebuild, and it spans most of the systems a working engineer treats as black boxes: your own database, git, Docker, a programming language, a regex engine, an operating system, a web server, a shell, a text editor, a neural network, a blockchain, a BitTorrent client, and on through dozens more categories. Each entry points to one or more external tutorials, often in several languages, so you can usually find a walkthrough in a stack you already know. The implicit argument the catalog makes is that almost any tool you depend on is comprehensible if you build a small version of it, and that the inventory of “things you could demystify” is much longer than most engineers assume until they scroll it.
How to use it well
Because it is an index of third-party tutorials, getting value from it is about how you navigate it:
- Pick by curiosity, not completeness. The list is huge. Choose the one technology you most want to stop treating as magic and build that, rather than working top to bottom.
- Expect variable freshness and quality. These are community-contributed links to external tutorials of different ages and depths. Some are pristine, some have bit-rotted; treat a broken or dated tutorial as a list entry to skip, not a dead end.
- Mind the missing difficulty signal. The single most-discussed open issue asks exactly this: what are the difficulty levels, and what should you know first? The list does not grade entries, so gauge prerequisites yourself before committing to one.
There is nothing to install. You browse the README, follow a tutorial link, and build the thing in your own environment.
The commercial context worth knowing
The repository is maintained by CodeCrafters, which sells paid, interactive “build your own X” courses with automated test suites and real-time feedback. That does not diminish the free list, which stands on its own and predates none of its usefulness, but it is worth knowing the relationship: the open index is also the top of a funnel toward a paid product for the same idea executed as a guided, graded experience. Whether you want the free do-it-yourself path or the paid guided one is a real choice, and the list makes the free path entirely viable.
build-your-own-x versus a structured course
| build-your-own-x | ai-engineering-from-scratch | |
|---|---|---|
| Stars | 513,858 | 30,905 |
| Form | curated index of external tutorials | original structured curriculum |
| Scope | general technologies | AI engineering specifically |
| Path | self-directed, pick one | linear phases and lessons |
Counts are from GitHub as of June 2026. The instructive contrast is with ai-engineering-from-scratch, which applies the same build-it-yourself philosophy but as an original, linear curriculum for one domain rather than a curated index across many. Build Your Own X is broader and self-directed; the other is narrower and structured. Choose by whether you want a map of where to learn anything, or a guided path through one field.
Related
For the same rebuild-it-to-learn-it philosophy applied specifically to AI, see ai-engineering-from-scratch. For what else is climbing on GitHub, see the daily digest and the weekly report.
FAQ
Is this a course? No. It is a curated index of external tutorials for recreating technologies from scratch. You follow the links and build the projects yourself.
Why so many stars? The rebuild-to-understand approach is widely loved, and this is the canonical list for it, accumulated since 2018.
Are the tutorials kept up to date? They are community-contributed external links of varying age and quality. Expect some to be dated or broken, and skip those.
Who maintains it? CodeCrafters, which also sells paid interactive versions of the same idea. The free list stands on its own.