00
Home
01
Built
02
Teardowns
03
Writing
04
About
05
Just out
한국어
GitHub ↗
03 Writing
Verdicts
3 posts
The space between PASS and FAIL — what counts as evidence, and how much a machine may decide alone.
Measurement
1 post
Where the time actually went, and how to keep estimates and measurements apart. And the work of deciding what to compare a number against.
Running agents
2 posts
The other side — loops, guards, and the things I decided not to automate.
Choosing
2 posts
Deciding whether to bring someone else's work in. What counts as grounds for rejection, and when a rejection gets reopened.
03
Writing
Notes that outgrew a commit message.
Pick a project.
Everything at once
Verdicts
I did not touch the
threshold
Is it broken, or not there
yet?
"Probably right" is not a
verdict
Measurement
All four times, the numbers
were right
Running agents
Agents are quiet when they
fail
Unattended regression is
built by recovery, not
speed
Choosing
Forty rejections, and the
reason was rarely the tool
You don't have it if it has
never run
Writing
Everything in this section, down to the piece. The columns walk it one step at a time; this is the same thing on one page.