Jul 15, 2026
Shopify said yes.
Seventh build log. Two days after I wrote about submitting Kōkua and not knowing if it'd get approved, Shopify approved it. It's live on the App Store.
Build Log
What actually happened, in public.
What shipped, what Claude built versus what the operator built, real hours saved, real mistakes and the guardrails added after. First-person, not a highlight reel.
Jul 13, 2026
The filter benched my fun ideas. I built the one it kept.
Sixth build log. I ran a batch of ideas through my own filter. It benched the three I was excited about and kept an accessibility scanner I didn't want to build. Three days later it's a submitted Shopify app and a real company, and I still don't know if it'll get approved.
Jul 6, 2026
I built it twice. The fear changed shape the second time.
Fifth build log. Ten months, two attempts, three rebuilds, and a quiet quit in between, including the part I almost left out: neither time did I ask anyone in person. I posted instead, both times, and I'm still not sure if that's a distribution problem or something I'm avoiding.
Jul 5, 2026
I built the app in a day. Then I waited two weeks for an email.
Fourth build log. I checked my own commit history and found the real number: about 13 hours of work, under 24 hours start to finish. Then Apple's approval queue took over, and two and a half weeks later I'm still waiting.
Jul 4, 2026
The week I almost built the wrong thing
Third build log. I wired my AI team to text me for approvals, then nearly built a whole service to reply from my phone, before I caught that I was reinventing something Claude already ships. The catch is the point.
Jul 3, 2026
Standing up the Founding Four, for real
Second build log. I stood up the whole Founding Four on a real business: three AI cofounders, a chief of staff, an approval gate, all from the guide's templates. Then I turned it loose on a first batch. Here's what the process actually revealed, including the moment a cofounder refused to make something up.
Jul 1, 2026
Running the roadmap on my own agency
First build log. I ran Sparkfile's own 90-day roadmap on my own agency and it broke in a useful way: it assumed I was a product startup. Here's what the setup actually looked like, the pricing shift it forced, the two workflows that came out of it, and how it changed the roadmap.