If a model meets all success criteria but the infrastructure review reveals insufficient compute capacity for expected traffic, what should the go/no-go decision be?

Go, because the model performance is the most important factor

No-go, because the model is not accurate enough

Go, but with a plan to upgrade infrastructure later

No-go, because infrastructure readiness is a prerequisite for deployment

What is the main goal of robustness assessment in the context of model operationalization?

To verify that the model's code is free of bugs

To measure the model's training time on the given dataset

To ensure the model performs consistently when faced with variations in input data

To confirm the model's predictions are explainable to stakeholders

A churn-prediction model was trained on customer data from 2020-2023. The team evaluates it on a holdout set from 2023 and achieves strong performance. Before the go/no-go decision, the data science lead suggests evaluating the model on a dataset from the first quarter of 2024, which was not used in training. What is the primary purpose of this additional evaluation?

To replace the original holdout set because it is no longer valid.

To test the model's generalization to unseen, more recent data that reflects potential shifts in customer behavior.

To increase the size of the training dataset for the final model version.

To comply with a regulatory requirement that all models be tested on data from the current year.

What is the purpose of reviewing operational procedures during the go/no-go assessment?

To ensure monitoring, maintenance, and rollback plans are defined and feasible

To select the algorithm for the model

To decide the model's hyperparameters

To calculate the model's training loss

A team is reviewing a demand-forecasting model for operationalization. The model meets all performance targets. However, the operational procedures document does not include a rollback plan, and the monitoring dashboard is not yet configured to track prediction drift. What should the team do?

Approve the model but require the operations team to create the rollback plan within two weeks of launch.

Approve the model because performance targets are met and monitoring can be set up after launch.

Deploy the model and rely on the previous model's rollback plan, which is still available.

Issue a no-go decision until the monitoring dashboard is configured and a rollback plan is documented and tested.

PMICertified Professional in Managing AI (PMI-CPMAI)

Domain 4Objective 6

Verify Model Ready for Operationalization Go/no-Go Decision PMI-CPMAI Practice Questions (Page 1)

Part of the Manage AI Model Development and Evaluation domain, which accounts for 16% of the PMI-CPMAI exam. PMI does not publish an official question count, but from its 160-minute exam (~65–105 total, ~10–17 in this domain), expect 2–3 from this objective — we provide 21 practice questions to prepare you well beyond it. (estimate)

21questions here

5free pages

7concepts

16%of the exam

Questions 1–5

1foundation · easy
If a model meets all success criteria but the infrastructure review reveals insufficient compute capacity for expected traffic, what should the go/no-go decision be?
Select an answer first
2foundation · easy
What is the main goal of robustness assessment in the context of model operationalization?
Select an answer first
3application · medium
A churn-prediction model was trained on customer data from 2020-2023. The team evaluates it on a holdout set from 2023 and achieves strong performance. Before the go/no-go decision, the data science lead suggests evaluating the model on a dataset from the first quarter of 2024, which was not used in training. What is the primary purpose of this additional evaluation?
Select an answer first
4foundation · easy
What is the purpose of reviewing operational procedures during the go/no-go assessment?
Select an answer first
5application · medium
A team is reviewing a demand-forecasting model for operationalization. The model meets all performance targets. However, the operational procedures document does not include a rollback plan, and the monitoring dashboard is not yet configured to track prediction drift. What should the team do?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to PMI-CPMAI

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by PMI. “PMI-CPMAI” is a trademark of its owner, used for identification only.