Web, 3D, visuals & the in-betweenAvailable for freelance & collaborationBased in PolandWeb, 3D, visuals & the in-betweenAvailable for freelance & collaborationBased in PolandWeb, 3D, visuals & the in-betweenAvailable for freelance & collaborationBased in Poland
Case Study2026

Personal portfolio website

A personal portfolio website built to not work in hospitality anymore, change my life, showcase my skills and to have a better quality of life.

Role

fullstack, UX/UI

Stack

Next.jsReactTypeScriptTailwindVercel

The problem

Most portfolios I admired did one of two things really well. Either they were obviously a developer's site — clean, precise, slightly intimidating in the GitHub-y way — or obviously an artist's site, with the weird cursor and the moodboard energy. I am both, which sounds good in a cover letter and looks like a mess on a homepage.

My first attempt was the kitchen sink: Blender renders, analog photos, code projects, sound experiments, all jostling for attention on the same page. It was a disaster. It looked like I couldn't decide what I wanted.

Design decisions

The site is split into /work and /experiments, doing different jobs. /work is where the deployable, clienty things live — proper case studies, tech stack, the whole pitch. /experiments is where I get to be weird without it counting against me. A scroll-driven zine doesn't need to justify itself the way a SaaS tool does.

No shadcn anywhere on the case study pages. I love shadcn for actual products, but on a creative-agency-facing portfolio it reads like "I know how to ship a SaaS dashboard." That's not what I'm selling here. So the typography, the motion, the spacing, the hover states — all custom, all on me to get right.

The motion language is restrained on purpose. Sections fade up as you scroll, sticky labels anchor section titles, and that's most of it. When you can do GSAP and Three.js, there's a temptation to put GSAP and Three.js on every page. I wanted the case studies themselves to be readable first, expressive second. The expressive lives one click deeper, in /experiments, where it belongs.

The hardest part

The actual hardest part wasn't technical — it was deleting things. I had a Blender showcase page mostly built. I had a sound-design demo. I had a TouchDesigner experiment I was proud of. Cutting them out of the navigation felt like throwing away parts of myself.

What helped was reframing it: a portfolio isn't an inventory of everything I can do. It's an argument for one specific thing. The other skills can show up inside the work — in the textures, the typographic choices, the pacing — without needing their own front-page real estate.

What it taught me

That focus is more impressive than range. Studios don't hire multidisciplinary in the abstract; they hire frontend developers who happen to have weird useful side skills. The fine arts background still does the work I want it to — it shows up in the typography, the comfort with negative space, the willingness to sit on a problem until it feels right — but it does that quietly, in service of the main story.

Also: ship the imperfect version. There's an /experiments page with one and a half things on it. There are case studies still in draft. But it's live, and it's better than the polished-forever-unreleased version that lived in my head for six months.