I see two mistakes here. First, they mention that they wanted to wait for the senior dev, but in the end they dont wait for him/her. The Second mistake: > We s

I see two mistakes here. First, they mention that they wanted to wait for the senior dev, but in the end they dont wait for him/her. The Second mistake: > We sent a ticket to the infrastructure team to run it directly on our servers There are some issues with this if i understand it correctly. They run the database changes on all environments at the same time. Why they would want to do that i don't know. When deploying such changes you should always test the migrations and the systems locally (not always possible), and then step them through dev-stage-production environments, /through the CI/CD system/. She also makes a point of not wanting to run it using the CI/CD due to now having oversight, but i fail to see how handing the scripts to the ops team is in any way superior in this regard. In any case, not such a big catastrophe. Just rewind the changes, assuming they were professional enough to write undo-scripts.