In May I taught a hands-on Web Development course at my school, teaching the students how to build a static website, and in the last few classes I also used vibe coding to show them how to make pages with AI. After the course ended, a dozen or so students came to talk to me: they thought the AI was great and wanted to keep learning, and asked whether I could keep teaching them.
So I split the ones who stayed into a few groups and gave them a website as an assignment, and they got more and more into it, until the groups merged into one real web team building one real project, web-tutorial.com.
They knew their backend and database skills were thin, so when we started planning, we decided to do static pages only: HTML, CSS, JavaScript, no server logic, no login system. AI generated the pages the whole way through, with all of us designing the architecture up front. We hit plenty of pits. Every group had its own style, and for a while the site looked like several different websites stitched together; AI also forgets things, writing a component on Monday and rebuilding it under another name on Tuesday, or uploading to the server before anyone had tested it locally; and the translation part was a mess too, starting with DeepL and later moving to a local model running on my own machine… eventually those problems got solved one by one, and after all that stumbling we somehow ended up with more than 14,000 pages across more than 40 technologies, and the site started to take shape.
Honestly, why am I posting this? Because our site has never had much traffic. English and Chinese sites have too many giants and we can't outrank them, so we tried Arabic, Portuguese and Japanese sites, and traffic stayed low anyway. I asked people in the industry and I asked a few AIs, and the answers came back to the same things: the site never really did SEO, the multilingual setup wasn't layered properly from the start, and there are no decent backlinks. We're working on the SEO now; for backlinks we're going into developer communities. So this post is half an introduction, and half the assignment we set ourselves, hoping it brings in some traffic.
The twelve students have been learning and fixing things the whole way, and watching how excited they are about it got me excited too. I don't know how long we can keep this going, but I want to try to keep the project alive: it helps this group learn new skills, it stays for the next group, and most of all the content really is free. It's rough in places, but if you only speak Japanese, Portuguese or Arabic and you want to learn programming, it's all there and free, and maybe while you're reading the tutorials you can help me get a bit of traffic too. It helps you, and it helps me.
Top comments (2)
twelve students to 14,000 pages is a wild ratio. the multilingual part is doing a lot of the work there.
You're right, and I should have said it in the post. The 14,000 is five languages of roughly 2,800 pages each — so the real content is about 2,800 pages, mirrored into four other languages (some fully translated, some still catching up). The multiplier is real, but it's also the point: the same lesson in Portuguese is a different page for a different reader.
Thanks for reading it closely enough to catch that.