A great story which really resonated with me from the operations / reliability engineering perspective. It is very common for junior engineers to jump into an a

A great story which really resonated with me from the operations / reliability engineering perspective. It is very common for junior engineers to jump into an action in case of a big outage before really trying to understand the problem. One example I've seen multiple times was to deal with a repeatedly crashing process by restarting yet another process. In some cases, this may help, but in others it can make the situation much much worse. Instead, the advice I took to heart is to delay action until you have a mental model of the problem, and how the planned action would resolve it (and if possible, sense check it with another engineer). In the best case scenario, you would have a trusted (well-tested) mitigation procedure that would provide you with enough time to recreate the issue and test the proposed solution in a staging environment.