Abstract:
PagerDuty has always put all developers on-call. In the last few years, we have transitioned from a handful of developers, working on a monolithic Rails app, to dozens of engineers and a dozen internal services. While the technical architecture and delegation of development work had very natural divisions, we found that it was important to keep on-call responsibilities broad to cover more than the service-level divisions. This creates several challenges, one of which is keeping on-calls empowered and happy covering pages from services they are not actively developing.
This talk will focus on the motivations behind keeping broad on-call responsibilities, the operational tactics that equip on-calls with the breadth of knowledge necessary to maintain the services they respond to, and the cultural strategies for keeping your engineers’ urge to kill each other from rising.
Speaker:
Owen has been a developer at PagerDuty for two years working on its highly available platform for processing critical alerts. Today, he focuses on a reliable messaging service that ensure all calls, messages, and emails are timely received by customers.