Courseiva
Free · No account needed · No credit card

Google Professional Data Engineer Practice Test

747 questions with instant explanations, domain breakdown, and wrong-answer analysis. Built for the real exam.

Instant feedback after each answer
Full explanations included
Domain score breakdown
Real exam: 120 min
Pass mark: 720/1000

Sample questions with explanations

This is exactly what you see during practice — question, options, and a full explanation after you answer.

A data scientist wants to test a new model version on a small percentage of traffic before full rollout. Which Vertex AI feature allows this?

AA/B testing
Endpoint traffic splittingCorrect
CModel monitoring
DModel versioning with canary deployments

Vertex AI Endpoint traffic splitting allows you to route a specified percentage of inference requests to different model versions deployed on the same endpoint. This enables gradual rollout by directing a small fraction of traffic (e.g., 5%) to the new model while the rest goes t…Read full explanation

A machine learning team wants to deploy a new model version for canary testing, where only 5% of traffic is routed to the new version. Which Vertex AI endpoint configuration supports this?

AHave the client application randomly select which model to call with 5% probability.
BDeploy the new version to a separate endpoint and direct 5% of users via a load balancer.
Configure the endpoint with traffic split: 95% to old version, 5% to new version.Correct
DUse an A/B testing framework outside of Vertex AI to compare results.

Vertex AI endpoints natively support traffic splitting, allowing you to route a specified percentage of requests to different model versions deployed on the same endpoint. By configuring a traffic split of 95% to the old version and 5% to the new version, you can perform canary t…Read full explanation

Which TWO actions can help reduce prediction latency for a Vertex AI endpoint?

AIncrease the number of features
Optimize the model architecture to reduce sizeCorrect
Use a custom prediction container with optimized dependenciesCorrect
DUse a larger machine type with more vCPUs

Optimizing the model architecture to reduce size directly decreases the computational load during inference, which lowers prediction latency. Smaller models require fewer floating-point operations (FLOPs) per prediction, enabling faster response times on Vertex AI endpoints.

Untimed Practice

Answer at your own pace. Explanation and domain tag shown immediately after each answer.

Timed Practice

Countdown timer starts immediately. Results and domain scores shown at the end — just like the real exam.

Why practice here?

Full explanations on every question

Not just the right answer — you get exactly why each wrong option is wrong, so you learn the concept, not the answer.

Domain score breakdown

After each session see your score by exam domain so you know exactly where to focus study time.

100% free, forever

No subscription, no trial, no email wall. Start a session in under 10 seconds.

Exam-style questions

Scenario-based, precise wording, realistic distractors — written to match what you actually see on exam day.

← All PDE questionsPDE exam guideStudy guidePractice by domain