How can you ensure that the development team prioritizes service reliability over new feature development when a major incident causes the service to miss its SLO, given that the team deploys new releases multiple times a week for a large service with a well-defined SLO?
