Interesting reflections, but what about the situation when the application depends on a number of external services/APIs in other organizations, outside the control of the DevOps team in charge of the application itself? The degraded performance and scalability problems may be due the same type of problems in those dependencies. To coordinate mitigation efforts across involved organizations may be difficult. I would welcome thoughts and blog posts on these "real world" problems, where public services become increasingly more and more hybrid and complex in nature. A possible path may be to look into the field of Observability (across systems) and how AI and Machine Learning may be tools to look into. This will of course still need coordination across organizations and sharing of metrics, log data etc.