Workshop · 4 in the fire
Nothing is finished until
it has failed twice.
studio magenta's prototype workshop. We melt things down in public and cast them again — four experiments running, three of them still unshipped.
On the floor
Four experiments
- P-01 Spike
Miracle Alarm
An alarm you switch off by answering a question out loud.
- P-02 Concept
Rhythm Runner: Korean History
A rhythm game and a Korean history course, forced into the same mould.
- P-03 Concept
Synopsis Master
An LLM story generator organised around who, what, when, where, why and how.
- P-04 Live
Taskholder Desktop
The lab's first graduate: a desktop task manager built around projects, not lists.
How the lab runs
We ask whether it can be made before we ask whether it sells.
magenta.LAB is closer to a workshop than a company. We take things that should not go together, force them into the same mould, and publish whatever comes out.
Read the philosophyFrom the log
Latest notes
The lab is open
Four prototypes are on the floor, and the process behind them gets published — including the parts that failed.
The AlarmKit alarm keeps ringing after you open the app
We concluded iOS was impossible. Then the requirement changed, and the same platform turned out to be fine. The analysis was not wrong — it answered a different question.
5W1H as a prompt structure: an open question
Does constraining a model to who / what / when / where / why / how produce more usable story skeletons than a free-form prompt? We do not know yet. This is what it would take to find out.
Looking for people to experiment with.
Collaboration, feedback on a prototype, or just a question — all of it is welcome.