My studio's own site. The first version shipped with a Three.js scene that looked impressive and cost more than it earned. I removed it, rebuilt the visual design around typography and speed, and moved the whole site off legacy Hostinger hosting onto Vercel.
01
The Three.js hero pulled a heavy dependency into the critical path of a site whose entire pitch is performance. A studio selling fast frontends can't open with a slow one.
02
The site ran on Hostinger hosting that fit neither the framework nor the deploy workflow no preview deployments, no edge network, manual pushes.
03
The domain's records spanned Hostinger (registrar, email) and Cloudflare, with live email on the domain. The migration had to keep mail and the site up throughout.
04
Deleting the 3D hero left a visual void, rebuilding the site around typography and hierarchy required rethinking the entire visual language.
05
Shipping a site that claims fast performance demands proof not just desktop scores but real-world mobile metrics under throttled conditions.
Deleted the dependency. removed Three.js entirely rather than optimizing around it. The strongest performance optimization is the code you don't ship.
Design rebuild. I replaced the 3D hero with a typography-led design consistent with the studio's positioning.
Hosting migration. moved the site to Vercel: preview deployments per branch, automatic SSL, edge delivery.
DNS restructure. reworked records across Hostinger and Cloudflare, keeping email (MX, DKIM, SPF, DMARC) and the live site running through the cutover.

Removing the 3D layer did more for mobile than any tuning could less JavaScript to download, parse, and execute on the devices and networks where it hurts most.

A lighter, faster studio site on modern infrastructure and a demonstration of the judgment call that matters more than any framework skill: knowing what to delete.
If this is the kind of work you need — performance, architecture, or a rebuild — a short call is the fastest way to find out whether I can help.