It reemphasizes the question of importance. Would a user accept their data needing a AI implementation of a ("manual") migration and their flow completely chang

It reemphasizes the question of importance. Would a user accept their data needing a AI implementation of a ("manual") migration and their flow completely changing? Does reliability to existing users even matter in the companies plans? If it isn't a product that needs to solve problems reliably over time then it was kind of silly to use a DBA that cost twice the Backend engineer and only handled the data niche. We progressed from there or regressed from there depending on why we are developing software.