Field notes, research, playbooks, and lessons. One place to follow the threads that matter to your work.
A signed macOS process, green tests, and an idle run loop did not prove that the user had a usable app. The launch investigation that separated disk work, window state, and observation error.
A Parley notebook reported three landmark architectures as broken on cross-signer ASL. A warmup and a gradient clip brought all three back, and two matched the best model. The ranking had measured my training recipe, not the models.
We trained seven landmark architectures three times each. Three of them worked on one seed and collapsed to near-random on the others. A single-seed comparison would have called two of those collapses a result.
Our sign model averages 42% across signers. That average hides a range from 26% to 64% — and the thing that decides where a person lands is not the signs they make, it is who they are.
Our best landmark-only sign model scores 45% on signers it has never seen. The field routinely reports numbers twice that high. The lower number is the honest one, and it is the one we publish.
I keep a running catalog of how hearing-led sign-language AI fails. It is not a list of other people's sins. It exists so Parley can catch itself the moment it starts to look like one of them.
Three sign-recognition models recovered when the training recipe changed. A closer look at what model comparisons actually measure.
An audit of sign recognition on unseen signers, with multiple seeds and explicit limits on what the results establish.
Why one average accuracy score can conceal very different experiences across 21 signers.
Pre-work contracts co-signed by builder and evaluator. Eliminates close-enough shipping. 48 contracts shipped.
The discipline behind Parley's notebooks: question-first contracts, signer-holdout splits, multi-seed floors, failure-modes-first, and a Deaf-community honesty checklist. The rules that make a 45% you can trust beat an 85% you can't.
Five fields that turn fuzzy asks into concrete renders.
How to add work to a contract without losing rigor. When a re-sign is required.
The signs that your session is gradually losing accuracy. Restart hygiene that compounds.
