The product
Three levels — basic, middle and advanced — each built from numbered modules with a stated learning outcome and a knowledge quiz. Level one starts at the ethos and history of Bitcoin and a hands-on wallet workshop; level three ends in DeFi architecture, AMM mechanics and yield strategy. Around the curriculum sit educator profiles, enrolment, news and the legal pages a regulated-adjacent programme needs.
It is a learning-management platform, not a brochure: modules, learning outcomes and quizzes are structured records, enrolment and payment run through it, and the curriculum can be extended without anyone editing markup. That is also what makes the translation problem below a software problem rather than a copywriting one.
The programme overview and the entry point into the three levels.
Eleven locales, and why that number is the whole story
Croatian, English, German, Spanish, French, Slovenian, Macedonian, Albanian, Italian, Polish and Serbian. Each served under its own URL prefix, so a link shared in one language stays in that language and search engines can index each market separately.
Translating marketing copy eleven times is tedious. Translating a curriculum eleven times is a different problem: the module numbers, the learning outcomes and the quiz content all have to stay aligned across every locale, or a student in one language quietly gets a different course from a student in another. That constraint is what makes this a software problem rather than a translation invoice.
What we take from it
Blok3 and Tuatara Sailing are both ours, and they are very different businesses that share one shape: the operator speaks one language and the customer speaks another, and the product has to be correct in all of them at once. That is where an AI-first process stops being a talking point and starts being an economic argument — the eleventh locale costs almost what the second one did, because the checking is automated rather than trusted. We run that cost ourselves before quoting it to anyone else.