Why I build my frameworks in the open
Open source is not charity. For the kind of work I do, it is the fastest way to build trust, taste, and better software.
Most of the infrastructure I build — agents, voice transport, headless commerce — lives in public repositories. People assume that is generosity. It is not. It is the most selfish, effective way I have found to build good software.
Open code forces honest design
A private repo lets you hide your shortcuts. A public one does not. When anyone can read the commit, the “I’ll fix it later” branch never gets written, because later never comes and everyone can see it. Building in the open is a standing review from strangers I respect. That pressure makes the architecture better before the first user ever shows up.
Trust is the real product
When a founder hands me their idea, they are taking a bet on judgment they cannot yet verify. A pitch deck does not settle that. Readable, working code does. “Here is a framework I built and maintain, go read it” is a stronger claim than any list of adjectives on a résumé.
The loop gets shorter
Open work compounds. A pattern I solve once in kuralle-agents shows up again in a client build a month later, already battle-tested. The frameworks are not a side project that competes with paid work — they are the reason the paid work ships in days instead of weeks.
Build the thing in the open, and the open thing builds you back.
That is the whole strategy. This post is the first of many — notes from the build, written while the paint is still wet.