Blog
Notes on the gap between what a meeting decides and what actually gets done — integration guides, workflow write-ups, and the occasional note on how this thing is built.
- ·Engineering
Prefer flagging over guessing
An AI that admits it does not know who owns a task is more useful than one that picks someone plausible. Why the gaps in an extracted list are a feature, not a shortfall.
- ·Workflow
Why meeting action items never get done
Your notetaker captures every commitment and the work still slips. The gap is not detection — it is the twenty minutes of retyping nobody has after a call.