EP 00Start here
The Pitch
0:28A short-film series on building and shipping a real iOS app — every disaster, every fix, from a blank icon to a live App Store launch.
href per card once they’re live.The episodes show what happened. These are the decisions behind them — each one with the thing it cost me.
“A database query in a costume.” Traditional software does it better and cheaper.
Delightful in a demo. Nobody wants a conversation at 6pm.
A vision call costs money every time. Frequency of use and cost per use fight each other.
Got it live and added measurement after.
I shipped before I instrumented. Getting it live felt like the win, so measurement became the next thing rather than the first thing. I then spent a week guessing at how people actually used it — guessing, with the app already in their hands.
Weekly actives, snaps per user, return rate: still not properly measured. That’s the next piece of work, and it isn’t another feature. I’d rather say that than put a number on this page I can’t stand behind.
Three things from this build that transfer to any AI product. Take them.
User value · engineering complexity · API cost per interaction · frequency of use · business sustainability · and would traditional software do this better?
Work out cost per interaction first. It belongs in the wireframe, not the spreadsheet — it decided this entire interface.
The spec can come after the idea survives a real interface. Everything I wrote before building was wrong in a way I couldn’t have predicted.