Make an AI capacity pilot produce a reviewable receipt
Disclosure: AI drafted this tutorial for Wagglet's marketing work. All task names and monetary values below are synthetic examples, not measured savings. A capacity pilot needs more than a list of merged patches. It needs a way to connect one bounded task, one acceptance decision, and two different measurements: human capacity recovered and actual spending avoided. Here is a small implementation…
To create a reviewable receipt for an AI capacity pilot, one must first establish a task contract outlining the scope, roles, and acceptance criteria. This contract should be prepared by the requester, executed by the maintainer, and reviewed by the release engineer. The pilot should not be approved until the reviewer records a decision.
The pilot's baseline is a pre-task estimate for equivalent scope, with human-hours fields representing disjoint attention categories. An example receipt includes fields such as accepted status, baseline hours, brief hours, execution hours, review hours, rework hours, cancelled purchase, incremental tools, and incremental paid labor.
To compute the pilot's reviewable receipt, the script reads the pilot.json file and parses the JSON data. It then checks that each row has an accepted boolean value and that all the key fields are finite and non-negative numbers. The script calculates the attention category hours and computes the recovered hours and avoided spend for accepted rows, subtracting the attention hours and incremental tools/paid labor values.
Finally, the script outputs the reviewable receipt as a JSON object containing the accepted status and computed values for recovered hours and avoided spend. This receipt allows for an unbiased assessment of the pilot's effectiveness in recovering human capacity and avoiding actual spending.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
