VITON¹³ SCHOOL: a calm course landing page in five languages, with a live lesson on a laptop photo

VITON¹³ SCHOOL is the online school inside the VITON¹³ ecosystem, at school.viton13.com. It teaches three applied programmes — Artificial Intelligence, Design in Artificial Intelligence and Website SEO — priced in roubles at 1,300, 2,600 and 3,900 ₽.

The school needed a front page that explains three courses and one method without shouting, shows what a lesson looks like before anyone pays, and gives the same facts in English, Russian, Spanish, French and Hindi.

One landing page in five editions: English without a prefix, then /ru, /es, /fr and /hi, linked with hreflang. It runs as a sequence: a first screen with a ticker of the school’s thirteen skills, three course showcases with a manifesto card between them, counters, course cards with prices and checkout buttons, the four-step method, a three-question course planner, admission, news, guides, questions and answers and a waiting-list form.

The look follows calm AI product sites such as sanalabs.com: one grotesk, Inter Tight, set mostly at weight 400, black pill buttons, wide grey planes and one large still life per course. The six photographs were generated for the school with GPT Image 2.5 in Higgsfield.

In the Artificial Intelligence showcase the laptop screen is not part of the picture: we measured its position on a chroma-green original of the photo and placed a live HTML lesson over it, with a module list, the current step, a line for asking the mentor and a progress bar. Four tabs change what the screen shows; they advance every 6.5 seconds while the section is in view, stop as soon as a visitor picks one and stay still when the system asks for reduced motion. The design and SEO showcases show the same kind of step on a floating panel.

The landing is part of VITON¹³’s Next.js application and has its own subdomain. Course names, prices and modules come from one data file, so the counters, cards, menus and footer all read the same numbers; prices are formatted for each language, and the fine print says that nothing is charged before the school confirms a start with the learner.

The landing is live at school.viton13.com. On 27 September 2026 all five editions answered with status 200, and the school’s sitemap listed 30 URLs: in each language the landing, the news page, the guides index and three guides. We rendered all 30 at 390 px: each has one H1, a self-referencing canonical URL, hreflang to the five editions and x-default, and Open Graph; none scrolls sideways, and 30 internal link targets returned no errors.

That day we ran Lighthouse 13.4 three times on the home page for each form factor. On a simulated phone, performance scored 87, 93 and 93, with the largest contentful paint at 3.1 s in the median run, 21 ms of total blocking time and no layout shift. On desktop, performance was 100 in every run and the largest paint came at 0.8 s. Accessibility is 97, best practices 100 and SEO 100 on both; the whole page, photographs included, weighs about 0.8 MB on a phone and 1.0 MB on desktop.

This is VITON¹³’s own school, not a client project. The case shows how the page is built and how fast it loads, not how it works for learners: we report no enrolments, waiting-list sign-ups, sales or traffic.

The photographs are AI-generated and staged: nobody in them is a VITON¹³ student, teacher or customer, and the lesson on the laptop screen is sample content, not a learner’s work. The landing gives the images descriptive alt text but no visible AI label yet, so we state it here; the guide cards lower on the page use licensed Unsplash photographs. Lighthouse is a lab test from one MacBook on one day, not field data from real visitors, and the layout follows a genre set by another company’s product site; the copy, photographs and code are ours.

How to check a web studio’s portfolio →