Robin Cannon

Robin Cannon

Home
Field Notes
Signals
Shiny Toy Robots (Fiction)
About
The raid
Not here. That doesn’t happen here.
Nov 20
The AI PM is a decision-rights job
Agents are taking on product work. The PM ladder needs to say who decides, who learns, and who answers for the outcome.
Nov 17
Eighteen months of noodles
Extra scallions. Zero defects.
Nov 13
The skill I trust and cannot prove
I tried to prove my deep research skill beat the bare model. I ended up with disagreement to learn from instead.
Nov 10
Beneath the sea
A place where someone leaves the lights on.
Nov 6
Who holds memory holds truth
I built a memory of decisions for Claude Code. Now it works across multiple agents.
Nov 3

October 2026

Unable to find signal [Part 3: Dead calm]
Connection established.
Oct 30
I can dress myself from the command line (and you probably can’t)
What my wardrobe, an experiment, and a thirty-year-old courtroom scene can tell us about trusting AI.
Oct 27
Poke it with a stick
The agent has a hundred days of money left. It isn’t doing anything with them.
Oct 25
Unable to find signal [Part 2: Same boat]
Conditions are much improved.
Oct 23
Who owns the boring update?
AI makes creation cheaper. And maintenance more important.
Oct 20
Unable to find signal [Part 1: Off-season]
He wasn't feeling well. That part was true.
Oct 16
© 2026 Robin Cannon · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for writing, podcasting, and media founders.