Chaos Monkey
jargoncloudops culture
Netflix's tool that randomly kills production instances, and the name for deliberately injecting failure to prove resilience.
Built during Netflix's early-2010s cloud migration and open-sourced in 2012, Chaos Monkey terminates random virtual machines during business hours so that engineers design for failure instead of hoping against it. It grew into the Simian Army and the discipline now called chaos engineering. The bet is simple: failures you schedule are cheaper than the identical failures you do not.