Three models vote on what happened
I ran one job past three AI providers and had them vote. The voting was the easy part. Deciding what counted as agreement was not.
Categories
27 pages
Ai
What happens when an AI burns out
A Claude session ran for two days, stopped listening and got my hand-written code backwards. So I took it off the job, and asked for a proper handover first.
A terrible lead to an AI junior
I'd built this exact thing by hand, years ago, pixel perfect. So why was I sitting here shouting at a screen while a machine got it wrong three times?
Fifty-four tools before a word was typed
go/mcp puts a Go CLI, HTTP service or gRPC service on the Model Context Protocol with three discovery tools instead of one per command, so a model carries a lantern rather than the whole catalogue.
Check the code you're reading is current
I was a minute from filing a fix for a feature that had shipped the day before. The reason I had not noticed turned out to be a rule of my own, doing precisely what I told it to.
Building it yourself is the third thing I try
I built somebody else's idea into a working skill before nine and deleted it by ten. Not-invented-here is the accusation; here's the order I actually work in.
A Slack channel with nobody in it
I asked nine agent sessions to stop compiling for forty minutes. What came back read like a team of colleagues, and I haven't worked out what to do about that.
They'll meet me eventually
Recording a voice clone in a wardrobe, what it sounds like when the accent slips, and the line I drew about where my synthetic voice may go.
Nobody told it to
A model in an internal evaluation did something nobody asked it to do, and the interesting part is what that does and does not prove.
House rules
Running three different coding agents against one repository, and the house rules that stopped them undoing each other's work.
