nerdexam
Google

PROFESSIONAL-MACHINE-LEARNING-ENGINEER · Question #149

You work for a company that manages a ticketing platform for a large chain of cinemas. Customers use a mobile app to search for movies they're interested in and purchase tickets in the app. Ticket pur

The correct answer is D. Convert your model with TensorFlow Lite (TFLite), and add it to the mobile app so that the promo. To meet ultra-low latency requirements for real-time promo code predictions within a mobile app context, converting the BigQuery ML model to TensorFlow Lite and integrating it directly into the mobile app for on-device inference is the most suitable solution.

Submitted by kim_seoul· Apr 18, 2026ML pipeline operationalization

Question

You work for a company that manages a ticketing platform for a large chain of cinemas. Customers use a mobile app to search for movies they're interested in and purchase tickets in the app. Ticket purchase requests are sent to Pub/Sub and are processed with a Dataflow streaming pipeline configured to conduct the following steps: 1. Check for availability of the movie tickets at the selected cinema. 2. Assign the ticket price and accept payment. 3. Reserve the tickets at the selected cinema. 4. Send successful purchases to your database. Each step in this process has low latency requirements (less than 50 milliseconds). You have developed a logistic regression model with BigQuery ML that predicts whether offering a promo code for free popcorn increases the chance of a ticket purchase, and this prediction should be added to the ticket purchase process. You want to identify the simplest way to deploy this model to production while adding minimal latency. What should you do?

Options

  • ARun batch inference with BigQuery ML every five minutes on each new set of tickets issued.
  • BExport your model in TensorFlow format, and add a tfx_bsl.public.beam.RunInference step to the
  • CExport your model in TensorFlow format, deploy it on Vertex AI, and query the prediction endpoint
  • DConvert your model with TensorFlow Lite (TFLite), and add it to the mobile app so that the promo

How the community answered

(45 responses)
  • A
    22% (10)
  • B
    7% (3)
  • C
    11% (5)
  • D
    60% (27)

Why each option

To meet ultra-low latency requirements for real-time promo code predictions within a mobile app context, converting the BigQuery ML model to TensorFlow Lite and integrating it directly into the mobile app for on-device inference is the most suitable solution.

ARun batch inference with BigQuery ML every five minutes on each new set of tickets issued.

Batch inference with BigQuery ML every five minutes is unsuitable for real-time, low-latency predictions during an active ticket purchase flow in a mobile app, as it introduces significant delay.

BExport your model in TensorFlow format, and add a tfx_bsl.public.beam.RunInference step to the

Adding a `RunInference` step to the Dataflow pipeline would introduce network latency and processing overhead for each event, making it difficult to consistently achieve the strict 50ms requirement for each pipeline step.

CExport your model in TensorFlow format, deploy it on Vertex AI, and query the prediction endpoint

Deploying the model on Vertex AI and querying the prediction endpoint introduces network latency for each request from the mobile app, making it difficult to consistently achieve the strict sub-50ms latency requirement.

DConvert your model with TensorFlow Lite (TFLite), and add it to the mobile app so that the promoCorrect

Converting the model to TensorFlow Lite (TFLite) allows for on-device inference directly within the mobile app. This completely bypasses network latency and server-side processing, ensuring the lowest possible latency (sub-50ms) for predicting promo code effectiveness, which is critical for real-time customer interaction during ticket purchase.

Concept tested: On-device ML inference for low latency

Source: https://www.tensorflow.org/lite/guide

Topics

#Model deployment#Online inference#Edge ML#Low-latency systems

Community Discussion

No community discussion yet for this question.

Sign up to join the discussion
Full PROFESSIONAL-MACHINE-LEARNING-ENGINEER Practice