Portfolio · 2026

I like turning ideas into real projects and learning through experience. Some I've launched, others I've stopped, and a few remain unfinished. This is a collection of what I've worked on, what came out of it, and what I've learned along the way.

PAGEFLIGHT SHIPRELAY KAIROLOCK ROLLING RECALL LUMEA NSBG SMMA UE5 GAME
Fig. 00Eight projects. Filled: built or launched. Hollow: stopped.

I notice problems,try to build what solves them,and learn from what happens.

Selected work

The code in these projects was written with AI coding tools. My part was deciding what to build, defining how it should work, directing the work and reviewing the results.

Nº 01Launched

PageFlight

pageflight.app (opens in a new tab)
Why
I wanted the experience of finishing a project and actually launching it.
Features
Enter a website URL, and PageFlight evaluates the site against predefined criteria and suggests areas for improvement.
Outcome
Launched and currently functioning as intended.
Fig. 01.1A real check of example.com, from entering the address to the report. Recorded September 2026.
Nº 02Discontinued

ShipRelay

Why
I was using two AIs for development—one to implement and the other to oversee the work. I built ShipRelay to reduce manual prompt transfers, address interruptions and unreliable AI output, and automate the deployment process.
Goal
Turn a given goal into a working product as quickly as possible, with minimal human intervention.
Outcome
I no longer use ShipRelay itself, but I still apply ideas I developed during and after the project to my development workflow.
Fig. 02.1A small to-do app that one ShipRelay run produced, started again from the saved files in September 2026.

ReproductionsThese ShipRelay images were recreated afterwards from the project's saved files, database and code. They are not original screenshots and differ from what ShipRelay actually looked like while it was running.

Nº 03Stopped

KairoLock

Why
I tried several app blockers to better control my distractions, but I kept finding ways around them. So I decided to build my own.
Goal
Create an app blocker that would be impossible to bypass.
Outcome
I stopped development after realizing I couldn't guarantee the level of blocking I wanted and doubting whether the product could compete on price.
Fig. 03.1The development build, recorded September 2026. It ran in a dry-run mode, so nothing was actually blocked — the app says so on screen.
Nº 04Launched

Rolling Recall

rolling-recall.pages.dev (opens in a new tab)
Why
I've always wanted to study in the most efficient way, so the same hours teach me more. I found the memorization method that works best for me and turned it into a website I use myself.
Goal
Since I had already built it, release it publicly in case it helps other people too.
Outcome
Launched. No response from users yet.
Fig. 04.1Adding a card and recalling it in both directions until it's done. Recorded November 2026; starts after sign-in, with the account email covered.
Nº 05Launched

Lumea

lumea.page (opens in a new tab)
Why
While typing one day, I thought it would be great if typing could always feel as good as possible, so I built this right away.
Goal
A website where writing or typing anything leaves you feeling as positive as possible.
Fig. 05.1From the home page to typing, then the settings and documents menus. Recorded November 2026; the account email in the settings panel is covered.

Other experiences

Nº 06Website built

NSBG

No Salt but Great

nosaltbutgreat.com (opens in a new tab)
Fig. 06.1Screen recording of nosaltbutgreat.com, October 2026.
Why
A family member asked me to build a website for their company.
Goal
Turn the brand's design direction into a web experience and interactions.
Outcome
The website is partially completed, with some work still unfinished.
Nº 07Discontinued

SMMA

Social media marketing agency

Why
I wanted to run a social media marketing business myself.
Goal
Find clients and set up a way to work with collaborators.
Outcome
I spent so much time perfecting small details that six months passed before I made meaningful progress in finding clients. I sometimes spent days writing a single cold email, eventually lost motivation, and stopped pursuing the business.
Nº 08Stopped

Unreal Engine 5 game

Why
I've long wanted to experience the atmosphere of a successful New Yorker's life through a highly realistic game, so I decided to try creating one.
Goal
Build a game that feels closer to real life than anything currently available.
Outcome
I stopped development after realizing how difficult it would be to achieve that level of realism on my own. I may revisit the idea in the future if I can secure the necessary funding.

What I learned

  1. 01

    Validate before building

    KairoLock and my UE5 project taught me to investigate technical limitations and the resources required before committing to ambitious ideas.

  2. 02

    Progress needs verification

    ShipRelay taught me that activity doesn't necessarily mean progress. I need to verify actual results rather than rely on what AI reports.

  3. 03

    Perfectionism can prevent execution

    My SMMA experience showed me how spending too much time on small details can prevent meaningful progress.

Long-term

So far I've mostly worked on software problems. Long term, I want to work on the physical limits of the human body — building the experience, track record and trust to help lead work on artificial organs and bodies.