series · 3 essays
Rewrites, Repriced
Every long-tenured engineering organization carries a modernization backlog it has learned to live with, because the full rewrite was priced as a career risk. These three essays argue that the price just changed, and that the change is measurable.
The argument runs in sequence. Legacy is not a moral failure; it is accumulated decisions that outlived their context. The rewrite taboo was a price tag in disguise, and agents that can port half a million lines in days rewrote the tag. What remains is an allocation question: at enterprise scale, the first modernization dollar should fund the answer key that lets a system grade its own replacement, not the port.
The supporting reads are the case studies: the same repricing observed at side-project scale, one rebuild at a time.
reading order
The essays
- 01Yesterday's Feature Is Today's Bug5 min read
The inheritance: every long-tenured organization carries decisions that made sense once and make everything harder now.
- 02Everyone Knows You Never Rewrite19 min read
The taboo breaks: the full rewrite was a price tag, not a principle, and half a million lines of Bun in eleven days changed the tag.
- 03The Modernization Backlog Just Got Repriced12 min read
The enterprise turn: systems that can grade their own replacement reprice the backlog, so fund the answer key before the port.
supporting reads
Around the spine
- Eleven Years, One Week, and an AI Co-Pilot: Rebuilding TravelTimes for iOS 26.4
Proof at side-project scale: six major features on a decade-old codebase in one week, and where the AI got it wrong.
- From iOS to Android in a Weekend: How I Used Claude Code to Build a Complete App Without Writing a Single Line of Code
The port as a benchmark: a production iOS app became a working Android app in eight hours.
- Native Was the Long Way Around. The Stack Wasn't the Point.
The discipline argument: rebuilding as native software demanded the same rigor agents need to be useful.