0 published runs

What these actually did.

Specs are easy to write and easy to believe. These are records of running them against real data, redacted — what the spec was given, what it returned, what it got wrong, and what a person had to fix before the output was usable.

Nothing published yet. This page stays empty until there is something real to put on it. Invented case studies would defeat the point of publishing them, so it waits.

Ran one at your firm?

Send what it returned and what it got wrong. Published with your name on it, or anonymized to your firm's profile if you would rather.

Get in touch