Why your agent's rework rate is an architecture problem, not a model problem

Which model is best for agents? Wrong question. Two teams can run the same model with the same prompts and see rework rates three times apart.