Optimizing foundation modelsFull notesSummaryRingkasanStoriesPracticeAgents7 exam-style questions on this lesson.Question 1 of 7What do agents add to a generative AI model?AAutomatic removal of bias from outputsBThe ability to take actions, not just produce answersCA larger training dataset for the modelDA lower cost for each token processedCheck answerQuestion 2 of 7An agent connects a model to a CRM platform and a service management tool. Which agent function is this?AContinuous pre-trainingBGuardrailsCIntermediary operationsDFeedback integrationCheck answerQuestion 3 of 7An agent processes a refund and updates the customer's account settings. Which agent function is this?AEmbedding generationBFeedback integrationCIntermediary operationsDLaunching actionsCheck answerQuestion 4 of 7An agent records the outcome of each action so the system can improve over time. Which function is this?AFeedback integrationBPrompt leakingCLaunching actionsDIntermediary operationsCheck answerQuestion 5 of 7In the telecom case, one agent saves data from each conversation into the enterprise database. What does this improve?AThe response rate of the satisfaction surveyBFuture RAG answers, because the knowledge base growsCThe model's weights, through retrainingDThe speed of the GPUs running inferenceCheck answerQuestion 6 of 7A user asks an assistant to calculate a total and then return the result formatted as JSON. What kind of task is this?AA single-step classification of the inputBA clustering task that groups similar inputsCA multi-step task that chains actions togetherDA batch transform job over a large datasetCheck answerQuestion 7 of 7In the telecom case, which agent tracks the 4-out-of-5 satisfaction goal?AThe agent that changes customers' plansBThe agent that adds conversations to the databaseCNone, because agents can't contact customersDThe agent that sends a survey at the endCheck answerRetrieval-Augmented Generation8 exam-style questions on this lesson.Evaluate Results6 exam-style questions on this lesson.