In designing a serverless ML system to enhance customer support tickets with relevant metadata before they are assigned to a support agent, a series of models are required to predict ticket priority, estimate ticket resolution time, and perform sentiment analysis. These models aim to equip agents with actionable insights for better decision-making. The tickets are general and do not contain domain-specific jargon. The architecture must be cost-effective, scalable, and comply with data privacy regulations. Given these constraints, which endpoints should the Enrichment Cloud Functions target? Choose the best two options if E is available, otherwise select the single best answer. | Google Professional Machine Learning Engineer Quiz