Extras · 1 of 4

A typical day

Start by checking what broke overnight and what alerts fired. Then a stand up meeting. Most of the morning is code: pipeline changes, infrastructure written as configuration files, a script to stop a problem recurring. Afternoons are reviews, helping a developer whose deployment is stuck, and slow work on a bigger migration. Roughly one week in four or five you carry the on-call phone, which can go off at 3am and does.