How should I use these Collaborating to manage data and models practice questions?

Read each scenario carefully and choose your answer before revealing the explanation. Then check why your choice was right or wrong. Repeat until the reasoning feels automatic.

PMLE · topic practice

Collaborating to manage data and models practice questions

Q: Can I practise just Collaborating to manage data and models questions in a focused session?

Yes — use the session launcher on this page to start a 10-, 20-, 30- or 50-question session drawn entirely from the Collaborating to manage data and models domain.

Practise Google Professional Machine Learning Engineer Collaborating to manage data and models practice questions — original exam-style scenarios with answer choices, explanations, and analysis of common mistakes.

Courseiva uses original exam-style practice questions designed for learning and revision. The goal is to understand the concepts, recognise exam patterns, and improve through explanations — not memorise copied exam dumps.

Reviewed byJohnson Ajibi· MSc IT Security

20 questionsDomain: Collaborating to manage data and models

Practice 10 questions Browse domain →

What the exam tests

What to know about Collaborating to manage data and models

Collaborating to manage data and models questions test whether you can apply the concept in context, not just recognise a definition.

How the topic appears in realistic exam-style scenarios.

Which detail in the question changes the correct answer.

How to eliminate plausible but wrong options.

How to connect the question back to the wider exam objective.

Watch out for

Common Collaborating to manage data and models exam traps

▸Answering from memory before reading the full scenario.
▸Missing a constraint such as cost, availability, security, scope or command context.
▸Choosing a broad answer when the question asks for the most specific fix.
▸Ignoring why the wrong options are tempting.

Practice set

Collaborating to manage data and models questions

20 questions · select your answer, then reveal the explanation

Question 1mediummultiple choice

Read the full NAT/PAT explanation →

A data science team uses BigQuery to store raw data and Vertex AI for model training. They want to ensure that only authorized users can access training data, and that model artifacts are automatically versioned and tracked. Which combination of Google Cloud services should they use?

Trap 1: Dataflow for data access control and Vertex AI Experiments for…

Dataflow is for data processing, not access control; Vertex AI Experiments is for tracking hyperparameters, not full model versioning.

Trap 2: Cloud Storage with bucket-level IAM and Cloud Build for versioning

Cloud Storage does not provide fine-grained access control for features, and Cloud Build is for CI/CD, not model versioning.

Trap 3: Cloud Composer for data access control and Cloud Source…

Cloud Composer is an orchestration tool, not for access control; Cloud Source Repositories is for code, not models.

Study all Collaborating to manage data and models common traps →

A
Dataflow for data access control and Vertex AI Experiments for model tracking
Why wrong: Dataflow is for data processing, not access control; Vertex AI Experiments is for tracking hyperparameters, not full model versioning.
B
Cloud Storage with bucket-level IAM and Cloud Build for versioning
Why wrong: Cloud Storage does not provide fine-grained access control for features, and Cloud Build is for CI/CD, not model versioning.
C
Cloud Composer for data access control and Cloud Source Repositories for model versioning
Why wrong: Cloud Composer is an orchestration tool, not for access control; Cloud Source Repositories is for code, not models.
D
Vertex AI Feature Store with access control and Vertex AI ML Metadata for model versioning
Vertex AI Feature Store provides controlled access to features, and ML Metadata tracks model artifacts and versions.

Full breakdown with real-world context →

Question 2hardmultiple choice

Read the full Collaborating to manage data and models explanation →

An ML team uses Vertex AI Pipelines to automate model retraining. The pipeline includes a step that queries BigQuery to create a training dataset. The team notices that the pipeline fails intermittently with a '403 Exceeded rate limits' error. What is the most likely cause and solution?

Trap 1: The training dataset is too large; partition the table and query…

The error is about rate limits, not data size.

Trap 2: The pipeline step timeout is too short; increase the timeout to 30…

Timeout increase does not resolve rate limit errors.

Trap 3: The SQL query is inefficient; rewrite it using materialized views

Inefficient queries cause timeout, not rate limits.

Study all Collaborating to manage data and models common traps →

A
The pipeline is issuing too many concurrent queries; use a BigQuery reservation to guarantee slot capacity
Reservations provide dedicated slots, avoiding API rate limits.
B
The training dataset is too large; partition the table and query only the latest partition
Why wrong: The error is about rate limits, not data size.
C
The pipeline step timeout is too short; increase the timeout to 30 minutes
Why wrong: Timeout increase does not resolve rate limit errors.
D
The SQL query is inefficient; rewrite it using materialized views
Why wrong: Inefficient queries cause timeout, not rate limits.

Full breakdown with real-world context →

Question 3easymultiple choice

Read the full Collaborating to manage data and models explanation →

A company stores training data in Cloud Storage and uses Vertex AI Training for model training. They want to implement a data validation pipeline to detect data drift before retraining. Which service should they use?

Trap 1: BigQuery ML

BigQuery ML is for building models, not drift detection.

Trap 2: Cloud Data Loss Prevention

DLP is for data privacy, not drift detection.

Trap 3: Dataflow

Dataflow is a processing engine, not a drift detection service.

Study all Collaborating to manage data and models common traps →

A
Vertex AI Model Monitoring
Vertex AI Model Monitoring can detect data drift by comparing distributions.
B
BigQuery ML
Why wrong: BigQuery ML is for building models, not drift detection.
C
Cloud Data Loss Prevention
Why wrong: DLP is for data privacy, not drift detection.
D
Dataflow
Why wrong: Dataflow is a processing engine, not a drift detection service.

Full breakdown with real-world context →

Question 4hardmultiple choice

Read the full Collaborating to manage data and models explanation →

A team uses Vertex AI Feature Store to serve features for real-time predictions. They notice that feature values are frequently updated from multiple source systems, leading to inconsistencies. They need to ensure that feature values are consistent across all serving endpoints. What should they do?

Trap 1: Use batch ingestion with weekly updates to reduce update frequency

Batch updates with long intervals increase the chance of serving stale features.

Trap 2: Increase the offline storage TTL to retain historical feature values

Retention does not affect consistency.

Trap 3: Implement a manual approval process for feature updates

Manual process is not scalable for frequent updates.

Study all Collaborating to manage data and models common traps →

A
Use batch ingestion with weekly updates to reduce update frequency
Why wrong: Batch updates with long intervals increase the chance of serving stale features.
B
Increase the offline storage TTL to retain historical feature values
Why wrong: Retention does not affect consistency.
C
Implement a manual approval process for feature updates
Why wrong: Manual process is not scalable for frequent updates.
D
Use a streaming ingestion pipeline with exactly-once semantics
Exactly-once streaming ensures each update is applied exactly once, maintaining consistency.

Full breakdown with real-world context →

Question 5mediummultiple choice

Read the full Collaborating to manage data and models explanation →

An organization uses Cloud Composer to orchestrate ML workflows. A DAG that triggers Vertex AI training jobs fails because the training job exceeds the 7-day maximum runtime. What is the best way to handle long-running training jobs in Cloud Composer?

Trap 1: Increase the DAG execution timeout to 14 days in the Airflow…

Cloud Composer has a 7-day limit for DAG runs, and increasing timeout may not be allowed.

Trap 2: Refactor the training job to run on Dataflow, which supports longer…

Dataflow is for data processing, not model training.

Trap 3: Set max_active_runs=1 in the DAG to prevent overlapping runs

This does not address the runtime limit.

Study all Collaborating to manage data and models common traps →

A
Increase the DAG execution timeout to 14 days in the Airflow configuration
Why wrong: Cloud Composer has a 7-day limit for DAG runs, and increasing timeout may not be allowed.
B
Use Vertex AI Pipeline to manage the training job asynchronously
Vertex AI Pipeline can handle long-running jobs independently of the DAG runtime.
C
Refactor the training job to run on Dataflow, which supports longer runtimes
Why wrong: Dataflow is for data processing, not model training.
D
Set max_active_runs=1 in the DAG to prevent overlapping runs
Why wrong: This does not address the runtime limit.

Full breakdown with real-world context →

Question 6easymultiple choice

Read the full Collaborating to manage data and models explanation →

A team wants to share a trained model with other teams within the organization. They need to provide access to the model artifact in Vertex AI Model Registry and ensure that only authorized teams can deploy the model. What should they do?

Trap 1: Grant the other teams access to the Cloud Storage bucket where the…

This bypasses Vertex AI's access control and may expose other artifacts.

Trap 2: Set the model to public in Vertex AI Model Registry

Public access is insecure and unnecessary.

Trap 3: Use Cloud Key Management Service to encrypt the model and share the…

KMS does not control deployment permissions.

Study all Collaborating to manage data and models common traps →

A
Grant the other teams access to the Cloud Storage bucket where the model is stored
Why wrong: This bypasses Vertex AI's access control and may expose other artifacts.
B
Set the model to public in Vertex AI Model Registry
Why wrong: Public access is insecure and unnecessary.
C
Use Cloud Key Management Service to encrypt the model and share the decryption key
Why wrong: KMS does not control deployment permissions.
D
Use IAM to grant the 'aiplatform.models.deploy' role to the other teams on the model resource
IAM roles provide fine-grained access control within Vertex AI.

Full breakdown with real-world context →

Question 7mediummultiple choice

Read the full Collaborating to manage data and models explanation →

A data scientist is using Vertex AI Workbench user-managed notebooks. They need to collaborate with a colleague on the same notebook. The colleague should be able to edit the notebook simultaneously. What should they do?

Trap 1: Store the notebook in Cloud Source Repositories and have the…

This allows version control but not real-time collaboration.

Trap 2: Share the underlying Compute Engine VM's SSH access with the…

This is insecure and does not provide simultaneous editing.

Trap 3: Export the notebook to Colab and share the link

Colab does not integrate with Vertex AI Workbench and may have compatibility issues.

Study all Collaborating to manage data and models common traps →

A
Store the notebook in Cloud Source Repositories and have the colleague clone it
Why wrong: This allows version control but not real-time collaboration.
B
Share the underlying Compute Engine VM's SSH access with the colleague
Why wrong: This is insecure and does not provide simultaneous editing.
C
Export the notebook to Colab and share the link
Why wrong: Colab does not integrate with Vertex AI Workbench and may have compatibility issues.
D
Share the notebook instance URL with the colleague; both can edit simultaneously
Vertex AI Workbench supports real-time collaboration through the same instance.

Full breakdown with real-world context →

Question 8hardmultiple choice

Read the full Collaborating to manage data and models explanation →

A team uses Vertex AI Pipelines with CustomJob components that pull training code from a Cloud Source Repository. The pipeline fails with a 'Permission denied' error when trying to access the repository. The service account used by the pipeline has the 'Source Repository Viewer' role. What is the likely issue?

Trap 1: The training code contains a dependency that is not available in…

The error occurs at the repository access step, not during execution.

Trap 2: The pipeline is running in a different project than the repository;…

Cross-project access is supported with proper IAM.

Trap 3: The repository URL is incorrectly formatted; use the SSH URL…

The error message indicates a permission issue, not a URL format issue.

Study all Collaborating to manage data and models common traps →

A
The training code contains a dependency that is not available in the custom container
Why wrong: The error occurs at the repository access step, not during execution.
B
The 'Source Repository Viewer' role is insufficient; the service account needs 'Source Repository Reader' or higher
Reader role allows cloning and fetching, while Viewer only allows browsing.
C
The pipeline is running in a different project than the repository; cross-project access is not supported
Why wrong: Cross-project access is supported with proper IAM.
D
The repository URL is incorrectly formatted; use the SSH URL instead of HTTPS
Why wrong: The error message indicates a permission issue, not a URL format issue.

Full breakdown with real-world context →

Question 9easymulti select

Read the full Collaborating to manage data and models explanation →

Which TWO statements about Vertex AI Feature Store are correct? (Choose 2)

Trap 1: Feature Store automatically applies feature engineering…

Incorrect: transformations must be implemented separately.

Trap 2: Feature Store can only store numerical features.

Incorrect: it can store various feature types.

Trap 3: Feature Store can only be used with Vertex AI models.

Incorrect: it can serve features to any model.

Study all Collaborating to manage data and models common traps →

A
Feature Store automatically applies feature engineering transformations.
Why wrong: Incorrect: transformations must be implemented separately.
B
Feature Store can only store numerical features.
Why wrong: Incorrect: it can store various feature types.
C
Feature Store can only be used with Vertex AI models.
Why wrong: Incorrect: it can serve features to any model.
D
Feature Store provides a centralized repository for feature data.
Correct: it centralizes features for reuse.
E
Feature Store supports both online and offline serving.
Correct: online for real-time, offline for batch.

Full breakdown with real-world context →

Question 10mediummulti select

Read the full Collaborating to manage data and models explanation →

Which THREE actions are best practices for managing ML models in production on Google Cloud? (Choose 3)

Trap 1: Manually tune hyperparameters for each retraining run.

Incorrect: automated tuning is more efficient.

Trap 2: Store all raw training data indefinitely for auditability.

Incorrect: data retention should follow policy, not indefinite.

Study all Collaborating to manage data and models common traps →

A
Manually tune hyperparameters for each retraining run.
Why wrong: Incorrect: automated tuning is more efficient.
B
Monitor model performance and data drift continuously.
Correct: monitoring helps detect degradation.
C
Use a central model registry for model governance.
Correct: registry provides control and auditability.
D
Version all model artifacts and training datasets.
Correct: versioning ensures reproducibility.
E
Store all raw training data indefinitely for auditability.
Why wrong: Incorrect: data retention should follow policy, not indefinite.

Full breakdown with real-world context →

Question 11hardmulti select

Read the full Collaborating to manage data and models explanation →

Which TWO factors should you consider when choosing between BigQuery and Cloud Storage for storing training data? (Choose 2)

Trap 1: The requirement for data encryption at rest.

Both services support encryption.

Trap 2: The need for fine-grained access control at the row level.

BigQuery supports row-level security; Cloud Storage does not.

Trap 3: The maximum size of the dataset (BigQuery limit 1 TB).

BigQuery has no such low limit.

Study all Collaborating to manage data and models common traps →

A
The format of the data: structured vs. unstructured.
Correct: Cloud Storage is better for unstructured data.
B
The need for SQL-based transformations and analysis on the data.
Correct: BigQuery supports SQL natively.
C
The requirement for data encryption at rest.
Why wrong: Both services support encryption.
D
The need for fine-grained access control at the row level.
Why wrong: BigQuery supports row-level security; Cloud Storage does not.
E
The maximum size of the dataset (BigQuery limit 1 TB).
Why wrong: BigQuery has no such low limit.

Full breakdown with real-world context →

Question 12hardmultiple choice

Read the full Collaborating to manage data and models explanation →

A financial services company uses Vertex AI to deploy multiple models for fraud detection. The ML team has set up a CI/CD pipeline using Cloud Build and Cloud Deploy. The pipeline builds a custom container with the trained model, pushes it to Artifact Registry, and deploys it to a Vertex AI Endpoint. Recently, a new regulation requires that all model deployments be audited and approved by the compliance team before going live. The compliance team wants to review the model's evaluation metrics and approve the deployment via a ticketing system. Currently, the CI/CD pipeline automatically deploys after the container is built. The team needs to implement a gating process without slowing down the development cycle. What should they do?

Trap 1: Use Cloud Composer to orchestrate the deployment and add a sensor…

This adds unnecessary complexity; Cloud Deploy is simpler and more appropriate.

Trap 2: Use Cloud Build's built-in approval gate feature to require…

Cloud Build does not have a built-in approval gate; Cloud Deploy does.

Trap 3: Store the model artifacts in Cloud Storage and have the compliance…

Manual deployment defeats the purpose of automation and is error-prone.

Study all Collaborating to manage data and models common traps →

A
Use Cloud Composer to orchestrate the deployment and add a sensor that waits for approval from the ticketing system via a custom operator.
Why wrong: This adds unnecessary complexity; Cloud Deploy is simpler and more appropriate.
B
Use Cloud Build's built-in approval gate feature to require compliance team sign-off before deployment.
Why wrong: Cloud Build does not have a built-in approval gate; Cloud Deploy does.
C
Modify the CI/CD pipeline to use Cloud Deploy's approval gate feature, requiring a manual approval from the compliance team before the deployment step.
Cloud Deploy supports manual approval gates integrated with the pipeline.
D
Store the model artifacts in Cloud Storage and have the compliance team deploy manually using the gcloud command.
Why wrong: Manual deployment defeats the purpose of automation and is error-prone.

Full breakdown with real-world context →

Question 13mediummultiple choice

Read the full NAT/PAT explanation →

A healthcare organization is building a machine learning model to predict patient readmission risk. They have sensitive data stored in BigQuery that includes protected health information (PHI). The data science team uses Vertex AI Workbench notebooks to explore the data and develop models. The organization's security policy requires that all PHI data must be encrypted at rest and in transit, and that access to the data is logged and audited. They also need to ensure that the data used for model training is de-identified to remove direct identifiers such as patient names and SSNs. The team wants to automate the de-identification process as part of the data pipeline. Which approach meets these requirements?

Trap 1: Enable Shielded VM on Vertex AI Workbench notebooks and use VPC-SC…

Shielded VM and VPC-SC provide security but do not de-identify data.

Trap 2: Use Cloud Key Management Service to encrypt the PHI columns in…

Encryption does not remove identifiers; the team would still see PHI after decryption.

Trap 3: Use BigQuery row-level security to mask PHI columns for the data…

Row-level security does not remove identifiers for training; it only masks at query time.

Study all Collaborating to manage data and models common traps →

A
Create a Dataflow pipeline that reads from the original BigQuery table, applies Cloud DLP de-identification transforms, and writes to a new BigQuery table. Grant the data science team access to the de-identified table.
Dataflow with DLP automates de-identification and creates a safe dataset.
B
Enable Shielded VM on Vertex AI Workbench notebooks and use VPC-SC to restrict data access.
Why wrong: Shielded VM and VPC-SC provide security but do not de-identify data.
C
Use Cloud Key Management Service to encrypt the PHI columns in BigQuery, and share the encryption key with the data science team.
Why wrong: Encryption does not remove identifiers; the team would still see PHI after decryption.
D
Use BigQuery row-level security to mask PHI columns for the data science team, and train the model directly on the original table.
Why wrong: Row-level security does not remove identifiers for training; it only masks at query time.

Full breakdown with real-world context →

Question 14mediumdrag order

Read the full Collaborating to manage data and models explanation →

Drag and drop the steps to deploy a trained TensorFlow model to Vertex AI Prediction in the correct order.

Drag steps to the numbered slots on the right, or tap a step then tap a slot.

Steps

Order

1Step 1

2Step 2

3Step 3

4Step 4

5Step 5

Question 15mediummatching

Read the full Collaborating to manage data and models explanation →

Match each regularization technique to its effect.

Drag a concept onto its matching description — or click a concept then click the description.

Concepts

Matches

Adds absolute value of weights to loss, induces sparsity

Adds squared magnitude of weights to loss, prevents overfitting

Randomly drops units during training to prevent co-adaptation

Stops training when validation performance stops improving

Increases training data diversity through transformations

Question 16mediummultiple choice

Read the full Collaborating to manage data and models explanation →

A team of ML engineers is collaborating on a project using Vertex AI. They want to ensure that only approved models are deployed to production. Which approach should they use?

Trap 1: Store all models in a Cloud Storage bucket and manually control…

IAM alone does not provide an approval workflow or version management.

Trap 2: Deploy models directly from training jobs to an endpoint without…

Direct deployment skips version management and approval.

Trap 3: Use Cloud Dataflow to transform raw predictions and then store them…

Dataflow is for data processing, not model management.

Study all Collaborating to manage data and models common traps →

A
Store all models in a Cloud Storage bucket and manually control access via IAM permissions.
Why wrong: IAM alone does not provide an approval workflow or version management.
B
Deploy models directly from training jobs to an endpoint without version tracking.
Why wrong: Direct deployment skips version management and approval.
C
Use Vertex AI Model Registry with version aliases to manage model versions and promote them after approval.
Model Registry provides version control, staging, and alias-based deployment.
D
Use Cloud Dataflow to transform raw predictions and then store them in BigQuery for analysis.
Why wrong: Dataflow is for data processing, not model management.

Full breakdown with real-world context →

Question 17hardmultiple choice

Read the full Collaborating to manage data and models explanation →

A company uses a Cloud Composer DAG to run a daily ML pipeline that includes Dataflow jobs and model training on Vertex AI. The pipeline frequently fails due to insufficient permissions when the Dataflow worker accesses data in Cloud Storage. What is the most efficient way to resolve this issue?

Trap 1: Grant the 'roles/storage.objectViewer' role to 'allUsers' on the…

This opens the bucket to the public, a security risk.

Trap 2: Use the Composer environment's service account for all pipeline…

Composer service account may lack data access or have overly broad permissions.

Trap 3: Move the Dataflow job to run after the pipeline so that data is…

Does not address the permission issue.

Study all Collaborating to manage data and models common traps →

A
Create a custom service account with required permissions and assign it to the Dataflow job.
Lets the Dataflow worker access the data securely.
B
Grant the 'roles/storage.objectViewer' role to 'allUsers' on the Cloud Storage bucket.
Why wrong: This opens the bucket to the public, a security risk.
C
Use the Composer environment's service account for all pipeline components.
Why wrong: Composer service account may lack data access or have overly broad permissions.
D
Move the Dataflow job to run after the pipeline so that data is already processed.
Why wrong: Does not address the permission issue.

Full breakdown with real-world context →

Question 18easymultiple choice

Read the full Collaborating to manage data and models explanation →

A data scientist wants to share a trained model with the team for review before deployment. The model is stored in Vertex AI Model Registry. What is the recommended way to grant the team read access to the model?

Trap 1: Grant the IAM role 'roles/aiplatform.admin' to the team members.

Admin role gives too many permissions, including deletion.

Trap 2: Export the model as a local file and share it via a shared drive.

Does not leverage Vertex AI's version control and is not a best practice.

Trap 3: Add the team members to the Cloud Storage bucket ACL with 'READER'…

Bucket ACL does not give access to Vertex AI model metadata.

Study all Collaborating to manage data and models common traps →

A
Grant the IAM role 'roles/aiplatform.admin' to the team members.
Why wrong: Admin role gives too many permissions, including deletion.
B
Export the model as a local file and share it via a shared drive.
Why wrong: Does not leverage Vertex AI's version control and is not a best practice.
C
Grant the IAM role 'roles/aiplatform.viewer' to the team members on the project.
This role allows viewing models in Vertex AI.
D
Add the team members to the Cloud Storage bucket ACL with 'READER' access.
Why wrong: Bucket ACL does not give access to Vertex AI model metadata.

Full breakdown with real-world context →

Question 19mediummultiple choice

Read the full Collaborating to manage data and models explanation →

Your team is using Vertex AI Feature Store for online predictions. You notice that feature values for some entities are missing in production, leading to failed predictions. Upon investigation, you find that the ingestion pipeline has been failing intermittently. What is the best immediate course of action to prevent prediction failures?

Trap 1: Set up monitoring alerts on the ingestion pipeline to get notified…

Alerting is useful but does not prevent current failures.

Trap 2: Change the prediction request to ignore missing features.

Model might require those features; ignoring them could cause errors.

Trap 3: Manually re-ingest all missing features by running the ingestion…

Temporary fix; intermittent failures will repeat.

Study all Collaborating to manage data and models common traps →

A
Configure default values for missing features in the feature store so that the model can fall back on them.
Ensures predictions can be made even when features are not available.
B
Set up monitoring alerts on the ingestion pipeline to get notified of failures.
Why wrong: Alerting is useful but does not prevent current failures.
C
Change the prediction request to ignore missing features.
Why wrong: Model might require those features; ignoring them could cause errors.
D
Manually re-ingest all missing features by running the ingestion pipeline again.
Why wrong: Temporary fix; intermittent failures will repeat.

Full breakdown with real-world context →

Question 20hardmultiple choice

Read the full Collaborating to manage data and models explanation →

A team of ML engineers is building a real-time fraud detection system. They use Cloud Pub/Sub to stream transactions, Dataflow for feature engineering, and Vertex AI to get predictions. They want to ensure that the data used for training matches the data used for serving to avoid training-serving skew. Which approach should they take?

Trap 1: Use a batch processing system for both training and serving to…

Batch processing is not suitable for real-time serving.

Trap 2: Implement separate feature engineering pipelines for training and…

Separate pipelines risk inconsistency.

Trap 3: Ensure that both training and serving read from the same Cloud…

Storage location does not guarantee same feature engineering logic.

Study all Collaborating to manage data and models common traps →

A
Use a batch processing system for both training and serving to ensure identical feature calculations.
Why wrong: Batch processing is not suitable for real-time serving.
B
Implement separate feature engineering pipelines for training and serving, but document them carefully.
Why wrong: Separate pipelines risk inconsistency.
C
Use Vertex AI Feature Store to store features computed during training and retrieve them in the serving pipeline.
Feature Store provides a consistent feature definition and computation.
D
Ensure that both training and serving read from the same Cloud Storage location.
Why wrong: Storage location does not guarantee same feature engineering logic.

Full breakdown with real-world context →

Continue with 20-question session →

Free account

Track your progress over time

Create a free account to save your results and see which topics improve across sessions.

Focused Collaborating to manage data and models sessions

Start a Collaborating to manage data and models only practice session

Every question in these sessions is drawn from the Collaborating to manage data and models domain — nothing else.

10 questions 20 questions 30 questions 50 questions

Browse all Collaborating to manage data and models questions →Mixed PMLE session

Frequently asked questions

What does the PMLE exam test about Collaborating to manage data and models?: Collaborating to manage data and models questions test whether you can apply the concept in context, not just recognise a definition.
How should I use these practice questions?: Select your answer before revealing the explanation. Then read why each option is right or wrong — this active recall approach builds retention far faster than re-reading notes.
Can I practise just Collaborating to manage data and models questions in a focused session?: Yes — the session launcher on this page draws every question from the Collaborating to manage data and models domain. Use a 10-question session first to gauge your baseline, then move to 20 or 30 once the weak spots are clear.
Where can I practise other PMLE topics?: Use the topic links above to move to related areas, or go back to the PMLE question bank to see all topics.
Are these real exam questions or dumps?: These are original practice questions written to test the same concepts the PMLE exam covers. They are not copied from any real exam or dump site.

Collaborating to manage data and models only

10 questions 20 questions 30 questions 50 questions

Mixed PMLE session

Track your progress

A free account saves results across sessions and highlights which topics need work.

Study resources

All PMLE questions Collaborating to manage data and models domain overview PMLE exam guide

Exam traps to avoid

▸Answering from memory before reading the full scenario.
▸Missing a constraint such as cost, availability, security, scope or command context.
▸Choosing a broad answer when the question asks for the most specific fix.
▸Ignoring why the wrong options are tempting.

Collaborating to manage data and models practice questions

What to know about Collaborating to manage data and models

Common Collaborating to manage data and models exam traps

Collaborating to manage data and models questions

A data science team uses BigQuery to store raw data and Vertex AI for model training. They want to ensure that only authorized users can access training data, and that model artifacts are automatically versioned and tracked. Which combination of Google Cloud services should they use?

An ML team uses Vertex AI Pipelines to automate model retraining. The pipeline includes a step that queries BigQuery to create a training dataset. The team notices that the pipeline fails intermittently with a '403 Exceeded rate limits' error. What is the most likely cause and solution?

A company stores training data in Cloud Storage and uses Vertex AI Training for model training. They want to implement a data validation pipeline to detect data drift before retraining. Which service should they use?

A team uses Vertex AI Feature Store to serve features for real-time predictions. They notice that feature values are frequently updated from multiple source systems, leading to inconsistencies. They need to ensure that feature values are consistent across all serving endpoints. What should they do?

An organization uses Cloud Composer to orchestrate ML workflows. A DAG that triggers Vertex AI training jobs fails because the training job exceeds the 7-day maximum runtime. What is the best way to handle long-running training jobs in Cloud Composer?

A team wants to share a trained model with other teams within the organization. They need to provide access to the model artifact in Vertex AI Model Registry and ensure that only authorized teams can deploy the model. What should they do?

A data scientist is using Vertex AI Workbench user-managed notebooks. They need to collaborate with a colleague on the same notebook. The colleague should be able to edit the notebook simultaneously. What should they do?

Which TWO statements about Vertex AI Feature Store are correct? (Choose 2)

Which THREE actions are best practices for managing ML models in production on Google Cloud? (Choose 3)

Which TWO factors should you consider when choosing between BigQuery and Cloud Storage for storing training data? (Choose 2)

Drag and drop the steps to deploy a trained TensorFlow model to Vertex AI Prediction in the correct order.

Match each regularization technique to its effect.

A team of ML engineers is collaborating on a project using Vertex AI. They want to ensure that only approved models are deployed to production. Which approach should they use?

A company uses a Cloud Composer DAG to run a daily ML pipeline that includes Dataflow jobs and model training on Vertex AI. The pipeline frequently fails due to insufficient permissions when the Dataflow worker accesses data in Cloud Storage. What is the most efficient way to resolve this issue?

A data scientist wants to share a trained model with the team for review before deployment. The model is stored in Vertex AI Model Registry. What is the recommended way to grant the team read access to the model?

Track your progress over time

Start a Collaborating to manage data and models only practice session

Related PMLE topic practice pages

Scaling prototypes into ML models practice questions

Automating and orchestrating ML pipelines practice questions

Collaborating within and across teams to manage data and models practice questions

Architecting low-code ML solutions practice questions