Tooling the rest of the team can run

What was at stake

Whether something was actually ready to ship lived in people's heads. Product and ops re-derived the same checks by hand every time, and the cost of that stays invisible right up until a guest hits the broken one.

The calls I made

1. 1Built internal tooling that audits readiness before release instead of after — the same checks, run the same way, by anyone.
2. 2Moved shipping a build into Slack: name a branch, it verifies, dispatches, and reports back in the thread with a way to cancel.
3. 3Wrote the checks so product and ops can run them without me in the loop.

What outlived it

• Release checks stopped depending on who remembered them.
• The team ships builds without asking an engineer.