PMLE Practice Test 36 — 15 Questions

Question 1

A company has a prototype ML model that achieves 85% accuracy on historical data. In production, accuracy drops to 70% after two weeks due to data drift. They need an automated retraining pipeline with minimal manual oversight. Which solution is most cost-effective?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Set up Vertex AI Model Monitoring to detect drift, which triggers a Cloud Function that submits a Vertex AI Training job with new data. It combines automated drift detection via Vertex AI Model Monitoring with a serverless retraining trigger (Cloud Function) that submits a Vertex AI Training job, minimizing manual oversight while only incurring costs when drift is detected. This avoids the expense of continuous retraining or always-on GPU instances, making it the most cost-effective solution for the described scenario.

Answer

Use Cloud Functions to trigger a Dataflow job that trains the model using custom containers

Answer

Deploy the model on a GPU-equipped Compute Engine VM and run retraining every time new data arrives

Answer

Schedule a weekly Cloud Composer DAG that runs a new training job with all available data

Question 2

A company needs to serve a large Transformer model (5 GB) with strict latency requirements (< 50 ms) and throughput of 1000 requests per second. The model is in SavedModel format. They are considering deployment options on Google Cloud. Which approach best meets these requirements?

Accepted Answer

Deploy on Vertex AI Prediction using a single high-memory VM with a GPU (e.g., n1-highmem-32 with A100).. A single high-memory VM with a powerful GPU (e.g., A100) can handle the model size and throughput with low latency, avoiding network overhead. Option B is wrong because Cloud Run does not currently support GPU instances effectively, and even with GPU, concurrency may not meet strict latency requirements. Option C is wrong because model parallelism across multiple GPUs on a single VM adds complexity and overhead that is unnecessary for a 5GB model that fits on a single high-end GPU. Option D is wrong because distributed serving across multiple VMs introduces network latency that would make it difficult to meet the 50 ms requirement.

Answer

Deploy on Cloud Run with a GPU-enabled instance and increase concurrency.

Answer

Deploy on Vertex AI Prediction using model parallelism across multiple GPUs on a single VM.

Answer

Deploy on Vertex AI Prediction using distributed serving with TensorFlow Serving and model sharding across multiple VMs.

Question 3

A data scientist wants to perform A/B testing between two model versions deployed on the same Vertex AI endpoint. They need to route 10% of traffic to the challenger model. Which approach should they use?

Accepted Answer

Deploy both models to the same endpoint and set traffic_split to 90 for champion and 10 for challenger. Vertex AI endpoints support traffic splitting directly, allowing you to route a percentage of requests to different model versions deployed on the same endpoint. By setting `traffic_split` to 90 for the champion and 10 for the challenger, the data scientist can perform online A/B testing without additional infrastructure. This is the simplest and most cost-effective approach, as it avoids managing separate endpoints or load balancers.

Answer

Use Vertex AI Experiments to compare models offline, then deploy the winner

Answer

Deploy the challenger model to a separate endpoint and use a load balancer to split traffic

Answer

Update the champion model with a new version and use model version aliases

Question 4

Your PyTorch training script uses DistributedDataParallel (DDP) across 4 vertices each with 4 GPUs (16 GPUs total). You submit a Vertex AI custom training job. How should you configure the worker pool spec?

Accepted Answer

Create one worker pool with 4 replicas, each with machine type having 4 GPUs. For DDP across multiple machines, use MultiWorkerMirroredStrategy equivalent in PyTorch: set replicas to 4, each with machine type having 4 GPUs. The TF_CONFIG env var is not needed; Vertex AI sets necessary environment variables for distributed training.

Answer

Create a chief worker pool with 1 replica (4 GPUs) and a parameter server pool with 4 replicas (no GPUs)

Answer

Create 4 separate jobs, each with 1 replica and 4 GPUs

Answer

Create one worker pool with 16 replicas, each with 1 GPU

Question 5

A startup wants to build a product recommendation engine without writing custom training code. They have user-item interaction data stored in BigQuery. Which Google Cloud service should they use?

Accepted Answer

BigQuery ML matrix factorization. BigQuery ML matrix factorization is the correct choice because it allows building a recommendation engine directly in BigQuery using SQL, without writing custom training code. It supports implicit and explicit user-item interaction data and provides built-in evaluation metrics, making it ideal for low-code ML solutions on existing BigQuery data.

Answer

Cloud Dataflow with ML APIs

Answer

Vertex AI AutoML Tables

Answer

Vertex AI Matching Engine

Question 6

An ML team is moving from a prototype Jupyter notebook to a production training pipeline. They want to ensure reproducibility. Which approach should they take?

Accepted Answer

Use a container with fixed dependencies and record hyperparameters.. Using a container with fixed dependencies and recording hyperparameters ensures that the training environment and configuration are captured, enabling exact reproduction. Option A is wrong because interactive parameter tuning is not reproducible—it introduces manual adjustments. Option C is wrong because exporting the notebook's output model directly lacks environment tracking and hyperparameter records. Option D is wrong because saving the notebook as a .py file does not capture the full environment or dependencies.

Answer

Use interactive parameter tuning.

Answer

Export the notebook's output model directly.

Answer

Save the notebook as a .py file.

Question 7

An ML engineer is using Vertex AI Vizier to tune hyperparameters for a custom training job. The training job takes 2 hours per trial. To speed up the process, the engineer wants to run 10 trials in parallel. What is the correct way to configure parallel trial execution?

Accepted Answer

Set the 'parallelTrialCount' parameter in the study configuration to 10. In Vertex AI Vizier, parallel trial execution is configured by setting the 'parallelTrialCount' field in the study configuration. The maxParallelTrials is not a direct field; instead, the StudySpec contains a StudyJobConfig with parallelTrialCount. Setting parallelTrialCount to 10 allows up to 10 trials to run concurrently.

Answer

Use the '--parallel-trials' flag in the gcloud ai hp-tuning-jobs create command

Answer

Set the 'maxParallelTrials' attribute in the HyperparameterSpec

Answer

Create a CustomJob with 'numTrials' set to 10 and 'parallel' flag

Question 8

You are monitoring a classification model that predicts loan default. The model was trained on data from 2020-2022. In 2023, the economic conditions changed, and the model's accuracy dropped significantly. Which monitoring approach would best help you detect this issue early?

Accepted Answer

Monitor the prediction distribution for significant shift from training distribution. Monitoring the prediction distribution for a significant shift from the training distribution directly detects changes in the model's output behavior, which is the earliest indicator of concept drift or data drift caused by economic changes. Unlike accuracy monitoring, this approach does not require labeled data, enabling real-time detection of performance degradation before ground truth labels become available.

Answer

Monitor the accuracy of the model on the latest batch of labeled data

Answer

Monitor feature distribution drift using KS test

Answer

Monitor the freshness of the training data

Question 9

A company is using Vertex AI Pipelines with reusable components. They observe that a component that performs hyperparameter tuning is failing intermittently with a 'ResourceExhausted' error. The component is configured with a small custom service account. What is the most likely cause?

Accepted Answer

The service account does not have sufficient quotas or permissions to create the required number of trials or workers. The 'ResourceExhausted' error in Vertex AI Pipelines typically indicates that the component is trying to create more resources (e.g., trials or workers for hyperparameter tuning) than allowed by the assigned service account's quotas or permissions. A small custom service account often has restricted quotas for AI Platform services, such as the number of concurrent trials or training workers, leading to this failure.

Answer

The component code has a bug causing infinite recursion

Answer

The KFP executor is not properly configured

Answer

The pipeline system memory is insufficient for the component

Question 10

A financial institution needs to deploy a fraud detection model with strict latency <100ms per prediction and high throughput (1000 predictions/sec). The model is a deep neural network. Which architecture on Google Cloud meets these requirements?

Accepted Answer

Use Vertex AI Prediction with autoscaling enabled and GPU machine types. Vertex AI Prediction with autoscaling and GPU machine types is correct because it provides low-latency online serving with autoscaling to handle high throughput (1000 predictions/sec) while keeping latency under 100ms. GPUs accelerate deep neural network inference, and autoscaling ensures resources match demand without over-provisioning.

Answer

Deploy the model on AI Platform Training with a single large VM

Answer

Deploy the model as a Cloud Function triggered by Cloud Pub/Sub

Answer

Use Vertex AI Batch Prediction with a fixed number of machines

Question 11

You are deploying a large deep learning model on Vertex AI endpoints. The model requires GPU acceleration and you want to minimize cold-start latency. Which TWO actions should you take? (Choose 2 correct answers)

Accepted Answer

Use a custom container that loads the model during startup.. Loading the model during container startup (e.g., in the Dockerfile's ENTRYPOINT or CMD) ensures that the model is already in memory when the first prediction request arrives, drastically reducing cold-start latency. This is a standard practice for Vertex AI endpoints where the container must be ready to serve immediately after scaling up.

Answer

Set minReplicaCount to 0 to allow scale-to-zero.

Answer

Increase maxReplicaCount to a high number.

Answer

Use batch prediction instead of online prediction.

Question 12

A data engineering team uses Dataflow for preprocessing and wants to integrate with Vertex AI Pipelines. They need to pass the preprocessed data location to the training step. What is the best practice?

Accepted Answer

Use PipelineParam to pass the output path. PipelineParam is the native mechanism in Vertex AI Pipelines (Kubeflow Pipelines SDK) to pass runtime outputs—such as a Cloud Storage path—between components. It creates a dependency graph that ensures the training step receives the exact output path from the preprocessing step, enabling dynamic, reproducible pipelines without hardcoding.

Answer

Store the path in Data Catalog

Answer

Use Cloud Pub/Sub

Answer

Write the output to a fixed Cloud Storage path and hardcode it in the pipeline

Question 13

A data science team is building a real-time feature engineering pipeline for ML model training and serving. They need to compute features from streaming data, store them for low-latency serving, and ensure consistency between training and serving. Which TWO Google Cloud services should they use?

Accepted Answer

Vertex AI Feature Store. Vertex AI Feature Store (A) is correct because it provides a centralized repository for storing, serving, and sharing feature data with low-latency online serving and batch serving for training, ensuring consistency between training and serving through point-in-time lookups and feature value time-stamping. Cloud Dataflow (D) is correct because it is a fully managed stream and batch processing service based on Apache Beam, enabling real-time feature engineering from streaming data with exactly-once processing semantics and automatic scaling.

Answer

BigQuery

Answer

Cloud Functions

Answer

Cloud SQL

Question 14

An application serving predictions from a Vertex AI endpoint receives many identical requests within a short time window. The team notices redundant computation and wants to cache responses to reduce latency and cost. What is the recommended solution?

Accepted Answer

Implement a cache layer using Cloud Memorystore for Redis, hashing prediction requests.. Vertex AI does not provide built-in request caching; instead, the recommended pattern is to implement an external cache like Cloud Memorystore for Redis. By hashing the prediction request payload and using it as a cache key, identical requests within the short time window can be served from Redis, eliminating redundant model inference and reducing both latency and cost.

Answer

Deploy the model on a larger machine type to handle duplicate requests faster.

Answer

Enable Vertex AI endpoint caching by setting the `enable_cache` flag.

Answer

Use Cloud CDN in front of the endpoint.

Question 15

A developer sees this error when calling the endpoint. What is the most likely cause?

Accepted Answer

The model is deployed but not yet serving. The question stem does not provide the exact error, but in Vertex AI, when an endpoint returns an error indicating that the model is not ready, the most common cause is that the model is deployed but still in a transient state (e.g., deploying, scaling, updating). Option B captures this scenario. Option C (no model deployed) would typically result in a different error, and other options are less likely. Thus, without the specific error text, this understanding points to the model being 'deployed but not yet serving' as the most probable cause.

Answer

The model is still in training

Answer

The endpoint has no deployed model

Answer

The request payload size exceeds limit