easyMultiple Choice
PDE Practice Question: A data science team has trained a TensorFlow…
A data science team has trained a TensorFlow model for image classification and wants to deploy it to production with minimal latency. They have already exported the model as a SavedModel directory. Which service should they use to create an online prediction endpoint?
⚠ Common exam trap
Many candidates confuse Vertex AI Endpoints with AI Platform Prediction (legacy) or think Cloud Functions can serve models, but the Google Professional Data Engineer exam tests that Vertex AI is the modern, fully managed service for online prediction with minimal latency, while the others are either deprecated or designed for different workloads.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Vertex AI Endpoints
Vertex AI Endpoints is the correct service for deploying a TensorFlow SavedModel to an online prediction endpoint with minimal latency. It provides managed, autoscaling infrastructure optimized for real-time inference, including GPU/TPU support, request batching, and automatic health checking, which are essential for production deployment.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Cloud Functions
Why it's wrong here
Cloud Functions runs short event-driven code snippets and cannot load a TensorFlow SavedModel to serve sustained low-latency image classification requests. It is tempting because it is serverless and simple to invoke, but it suits lightweight glue logic, not hosting a model artefact with predictable inference latency.
- ✓
Vertex AI Endpoints
Why this is correct
Vertex AI Endpoints deploys the SavedModel directory directly as an online prediction endpoint, serving TensorFlow models with low latency. It satisfies the minimal-latency constraint by hosting the model on managed infrastructure with autoscaling, avoiding the overhead of custom serving code or batch-only pipelines.
- ✗
AI Platform Prediction (legacy)
Why it's wrong here
AI Platform Prediction (legacy) is the older managed prediction service, superseded by Vertex AI, so it fails the requirement to build a current online endpoint from the SavedModel. It is tempting because it did host TensorFlow SavedModels for low-latency serving, and would have been correct before Vertex AI replaced it.
- ✗
Cloud Dataflow
Why it's wrong here
Cloud Dataflow is a managed Apache Beam runner for batch and streaming data pipelines, so it cannot host a SavedModel or expose an HTTP prediction endpoint. It is tempting because Dataflow handles large-scale data processing around ML workflows, but it belongs to preprocessing and feature pipelines, not online inference.
Go deeper
Related to this question
About these practice questions
One of 747 original PDE practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PDE practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PDE exam.