Writing
Write-ups on design engineering, AI products, 3D tooling and accessibility. Mostly about the load-bearing details: the bug that was not where I looked, the default that quietly did the opposite of what it promised, the unglamorous part that decides whether a thing actually works.
The first round of an AI Engineer interview was a recorded video with nobody on the other side. I sent my AI avatar to answer it, with a banner on every frame saying so, and approved each answer from a key on my desk. Why the disclosed version is the only one worth building.
I bought a Stream Deck, never installed its software, and turned eight small screens into the place where my AI agents ask permission, show their progress, and stop when I press a key. And the parts of macOS that fought back.
I suspected I rewrite the same ten things in every project. So I measured it, built the shared asset library that is supposed to fix it, and offered it to four fresh coding agents on tasks that matched it word for word. Zero pulls. Why consistency needs push, not pull.
My 3D compressor had been shipping uncompressed geometry since the day it launched. The encoder never loaded, the loader swallowed the 404, and every export still said "done". A story about tools that lie by omission, and the one assertion that would have caught it.
Every image model I ran carried "deformed" and "ugly" in its default negative prompt. For medical imagery that is not a quality setting, it is an instruction to delete the finding. What looked like a censored model was my own config fighting me.
Two days, three confident diagnoses, three fixes that all "worked". They worked because two thirds of the time anything works. What finally broke the loop was not a better theory, it was counting 39 launches already sitting in the logs.
I asked a local model who owns each domain my network talks to. It answered every one with total confidence, including the six domains I invented. The eval that mattered was not the real inputs, it was the ones that do not exist.
Building karaoke read-aloud in Lens - syncing a spoken word to its exact spot on screen, inside a document you can still edit, for readers with dyslexia.
Nothing under that tag yet.