being responsible for operating their services in production, 24/7. This includes development using shared infrastructure primitives, testing, on-call, reliability engineering, disaster recovery, defining SLOs, monitoring setup and alarming, debugging and performance analysis, incident root cause analysis, provisioning and deployment, etc. 「The human scalability of “DevOps”」 https://medium.com/@mattklein123/the-human-scalability-of-devops-e36c37d3db6a