AIF-C01 Question 125

Community Discussion

Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.

• 1

  Below is the correct answer: Realtime Batch Realtime

• 1

  In Scenario 1, the company's chatbot requires immediate predictions to accurately interpret user intent during interactions. Implementing real-time inference ensures that the model processes each user input instantaneously, providing timely and relevant responses essential for effective communication. For Scenario 2, the data processing job involves handling large volumes of text data during specific periods, such as weekends. Utilizing batch transform allows the system to process this data in bulk, optimizing resource utilization and efficiency without the need for immediate results. In Scenario 3, the engineering team aims to develop an API capable of swiftly processing small text inputs and delivering prompt predictions. Adopting real-time inference enables the API to handle each request as it arrives, ensuring low-latency responses critical for user-facing applications.

• 1

  real time batch real time chatbot (real-time, low-latency predictions): Real-time inference Data processing job (large datasets on weekends): Batch transform