Madman Studio is the public record around the work — what launched, what stayed online, what shut down, and what changed after each decision. It is not a polished trophy shelf.
Most product sites hide uncertainty. They show the best screenshots, the most stable promises, and the cleanest version of the story. That can sell one product, but it does not show how a builder makes decisions across many launches. This record keeps the rougher version visible: the live products, the dead ones, and the reasoning in between.
Why it exists
Product pages hide the failed experiments. Personal blogs hide the products. Madman Studio keeps both in the same room.
The point is a better signal than a portfolio gives. A portfolio flattens the story into selected wins. This record is built to show taste, judgment, shipping rhythm, recovery speed, and the willingness to write down what did not work. The graveyard is part of the product, not a side feature: if something died because the positioning was weak, the maintenance cost was wrong, or distribution never showed up, that belongs here. If something is still live, it carries a public state, not just a marketing sentence.
The Live and Graveyard shelves fill in on their own as I keep shipping and keep killing what no longer earns its keep. This page doesn’t try to catalog them — it just explains why they sit side by side.
Who’s behind it
I’m Marcus Ka, a Korean solo founder based in Seoul. I build and run small digital products by myself — from the first version to the boring maintenance after launch — and I’ve been doing it since 2015. Madman Studio is where I keep the evidence in the open: what I shipped, what I learned, and what I would do again. The fuller story lives on the about page; this is the short version.
How it works
A few rules keep the studio legible from the outside:
- Every product gets a state.
- Every meaningful launch gets a date.
- Every shutdown eventually becomes a postmortem.
- Notes hold the thinking around the work, not just announcements.
- The site stays readable even when the homepage looks loud.
Underneath the visual layer it is deliberately boring — Astro, MDX, content collections, served static and fast — so the record stays close to the work instead of becoming another dashboard to maintain.
The point
Keep launching, keep killing what no longer justifies maintenance, and keep enough evidence around that the next decision gets sharper. A visitor should be able to inspect the work — and the person behind it — without needing private context.