Exam Bundle

The frozen test set shows improved high-risk recall at an explicitly acceptable false-positive review volume. This is evidence for a different corrective change: Evaluate classification candidates using recall, precision-recall tradeoffs, and a decision threshold tied to default-review cost. That other change addresses: A loan dataset contains a rare default class, but a BigQuery ML model is selected using overall accuracy alone despite high false negatives. It does not validate the specific Stonewell Payments remediation for: A finance analyst has a trained BigQuery ML model but exports raw coefficients to a spreadsheet and manually reimplements inference with a mismatched schema. The necessary check instead is: Prediction output rows and model metadata match the SQL inference schema on a controlled sample without spreadsheet recoding.

A deliberately renamed required feature fails the preflight while a matching schema yields non-null results. This is evidence for a different corrective change: Validate prediction input column names and types against the registered BigQuery ML model signature before scoring. That other change addresses: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. It does not validate the specific Stonewell Payments remediation for: A finance analyst has a trained BigQuery ML model but exports raw coefficients to a spreadsheet and manually reimplements inference with a mismatched schema. The necessary check instead is: Prediction output rows and model metadata match the SQL inference schema on a controlled sample without spreadsheet recoding.

A controlled dataset with nulls produces complete training rows and a per-column missingness audit without label leakage. This is evidence for a different corrective change: Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. That other change addresses: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not validate the specific Stonewell Payments remediation for: A finance analyst has a trained BigQuery ML model but exports raw coefficients to a spreadsheet and manually reimplements inference with a mismatched schema. The necessary check instead is: Prediction output rows and model metadata match the SQL inference schema on a controlled sample without spreadsheet recoding.

Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. This action could be suitable for a different engineering incident: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not directly correct the reported Northgate Clinical Systems failure: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The discriminating acceptance check here is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Create a multi-series BigQuery ML forecast keyed by store identity and the event date rather than pooling locations into one series. This action could be suitable for a different engineering incident: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. It does not directly correct the reported Northgate Clinical Systems failure: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The discriminating acceptance check here is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Apply BigQuery ML K-means clustering to meaningful standardized unlabeled behavior features and inspect resulting segments. This action could be suitable for a different engineering incident: An operations team wants to segment unlabeled customer behavior yet supplies made-up class labels to a supervised classifier. It does not directly correct the reported Northgate Clinical Systems failure: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The discriminating acceptance check here is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Cluster assignments and separation diagnostics reveal coherent groups without requiring fabricated customer labels. This is evidence for a different corrective change: Apply BigQuery ML K-means clustering to meaningful standardized unlabeled behavior features and inspect resulting segments. That other change addresses: An operations team wants to segment unlabeled customer behavior yet supplies made-up class labels to a supervised classifier. It does not validate the specific Northgate Clinical Systems remediation for: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The necessary check instead is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

A controlled dataset with nulls produces complete training rows and a per-column missingness audit without label leakage. This is evidence for a different corrective change: Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. That other change addresses: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not validate the specific Northgate Clinical Systems remediation for: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The necessary check instead is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Forecasts retain store IDs and independent store-level holdout errors are available for inspection. This is evidence for a different corrective change: Create a multi-series BigQuery ML forecast keyed by store identity and the event date rather than pooling locations into one series. That other change addresses: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. It does not validate the specific Northgate Clinical Systems remediation for: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The necessary check instead is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Build a BigQuery ML time-series forecasting model with the timestamp and product series configured for seasonal forecasting. This action could be suitable for a different engineering incident: A marketplace predicts daily demand for each product but fits a static linear model without seasonal timestamps and misses weekly demand cycles. It does not directly correct the reported Stonewell Payments failure: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The discriminating acceptance check here is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. This action could be suitable for a different engineering incident: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not directly correct the reported Stonewell Payments failure: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The discriminating acceptance check here is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Prepare governed, task-labeled examples in BigQuery and use the documented Gemini tuning workflow where the model and region support it. This action could be suitable for a different engineering incident: A team requests domain-specific Gemini fine-tuning from curated BigQuery examples but repeatedly rewrites prompts instead of evaluating the supported tuning path. It does not directly correct the reported Stonewell Payments failure: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The discriminating acceptance check here is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

The held-out task rubric shows a measurable gain for the tuned candidate without contaminating validation examples. This is evidence for a different corrective change: Prepare governed, task-labeled examples in BigQuery and use the documented Gemini tuning workflow where the model and region support it. That other change addresses: A team requests domain-specific Gemini fine-tuning from curated BigQuery examples but repeatedly rewrites prompts instead of evaluating the supported tuning path. It does not validate the specific Stonewell Payments remediation for: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The necessary check instead is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Out-of-time product forecasts capture repeating weekly patterns and are compared against naive seasonal baselines. This is evidence for a different corrective change: Build a BigQuery ML time-series forecasting model with the timestamp and product series configured for seasonal forecasting. That other change addresses: A marketplace predicts daily demand for each product but fits a static linear model without seasonal timestamps and misses weekly demand cycles. It does not validate the specific Stonewell Payments remediation for: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The necessary check instead is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

A controlled dataset with nulls produces complete training rows and a per-column missingness audit without label leakage. This is evidence for a different corrective change: Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. That other change addresses: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not validate the specific Stonewell Payments remediation for: A regulated team moves data extracts to a different region for periodic ML scoring despite a residency requirement for the source dataset. The necessary check instead is: The job location and audit logs show local scoring, and no unauthorized cross-region export occurs.

Evaluate classification candidates using recall, precision-recall tradeoffs, and a decision threshold tied to default-review cost. This action could be suitable for a different engineering incident: A loan dataset contains a rare default class, but a BigQuery ML model is selected using overall accuracy alone despite high false negatives. It does not directly correct the reported Riverton Finance failure: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The discriminating acceptance check here is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Use a BigQuery ML recommendation approach such as matrix factorization with appropriate user, item, and interaction data. This action could be suitable for a different engineering incident: A media platform seeks implicit user-item recommendations but trains an ordinary numeric sales regression with user IDs as magnitudes. It does not directly correct the reported Riverton Finance failure: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The discriminating acceptance check here is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Remove future-known cancellation fields and rebuild features using only observations available at the prediction cutoff. This action could be suitable for a different engineering incident: A subscription model uses a churn-cancellation timestamp recorded after prediction time as a BigQuery ML input and reports implausible accuracy. It does not directly correct the reported Riverton Finance failure: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The discriminating acceptance check here is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

The frozen test set shows improved high-risk recall at an explicitly acceptable false-positive review volume. This is evidence for a different corrective change: Evaluate classification candidates using recall, precision-recall tradeoffs, and a decision threshold tied to default-review cost. That other change addresses: A loan dataset contains a rare default class, but a BigQuery ML model is selected using overall accuracy alone despite high false negatives. It does not validate the specific Riverton Finance remediation for: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The necessary check instead is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

A time-sliced audit proves no input timestamp occurs after the forecast cutoff and the leakage-driven score inflation disappears. This is evidence for a different corrective change: Remove future-known cancellation fields and rebuild features using only observations available at the prediction cutoff. That other change addresses: A subscription model uses a churn-cancellation timestamp recorded after prediction time as a BigQuery ML input and reports implausible accuracy. It does not validate the specific Riverton Finance remediation for: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The necessary check instead is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Held-out user-item recommendations rank relevant items more effectively than a non-personalized popularity baseline. This is evidence for a different corrective change: Use a BigQuery ML recommendation approach such as matrix factorization with appropriate user, item, and interaction data. That other change addresses: A media platform seeks implicit user-item recommendations but trains an ordinary numeric sales regression with user IDs as magnitudes. It does not validate the specific Riverton Finance remediation for: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The necessary check instead is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. This action could be suitable for a different engineering incident: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not directly correct the reported Brookfield Credit Union failure: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The discriminating acceptance check here is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Use a BigQuery ML recommendation approach such as matrix factorization with appropriate user, item, and interaction data. This action could be suitable for a different engineering incident: A media platform seeks implicit user-item recommendations but trains an ordinary numeric sales regression with user IDs as magnitudes. It does not directly correct the reported Brookfield Credit Union failure: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The discriminating acceptance check here is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Run BigQuery ML prediction against the governed input table with ML.PREDICT and the recorded feature mapping. This action could be suitable for a different engineering incident: A finance analyst has a trained BigQuery ML model but exports raw coefficients to a spreadsheet and manually reimplements inference with a mismatched schema. It does not directly correct the reported Brookfield Credit Union failure: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The discriminating acceptance check here is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Prediction output rows and model metadata match the SQL inference schema on a controlled sample without spreadsheet recoding. This is evidence for a different corrective change: Run BigQuery ML prediction against the governed input table with ML.PREDICT and the recorded feature mapping. That other change addresses: A finance analyst has a trained BigQuery ML model but exports raw coefficients to a spreadsheet and manually reimplements inference with a mismatched schema. It does not validate the specific Brookfield Credit Union remediation for: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The necessary check instead is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Held-out user-item recommendations rank relevant items more effectively than a non-personalized popularity baseline. This is evidence for a different corrective change: Use a BigQuery ML recommendation approach such as matrix factorization with appropriate user, item, and interaction data. That other change addresses: A media platform seeks implicit user-item recommendations but trains an ordinary numeric sales regression with user IDs as magnitudes. It does not validate the specific Brookfield Credit Union remediation for: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The necessary check instead is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

A controlled dataset with nulls produces complete training rows and a per-column missingness audit without label leakage. This is evidence for a different corrective change: Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. That other change addresses: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not validate the specific Brookfield Credit Union remediation for: A fraud team passes a renamed transaction field to ML.PREDICT and silently accepts null predictions from an incompatible scoring schema. The necessary check instead is: A deliberately renamed required feature fails the preflight while a matching schema yields non-null results.

Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. This action could be suitable for a different engineering incident: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not directly correct the reported Westhaven Marketplace failure: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. The discriminating acceptance check here is: Forecasts retain store IDs and independent store-level holdout errors are available for inspection.

Use a BigQuery ML recommendation approach such as matrix factorization with appropriate user, item, and interaction data. This action could be suitable for a different engineering incident: A media platform seeks implicit user-item recommendations but trains an ordinary numeric sales regression with user IDs as magnitudes. It does not directly correct the reported Westhaven Marketplace failure: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. The discriminating acceptance check here is: Forecasts retain store IDs and independent store-level holdout errors are available for inspection.

Use supported BigQuery ML model-evaluation and explanation functions at the appropriate global or local scope and distinguish contribution from causation. This action could be suitable for a different engineering incident: A customer support team cannot explain why an SQL-trained BigQuery ML model prioritizes a prediction and treats a coefficient printout as a complete local explanation. It does not directly correct the reported Westhaven Marketplace failure: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. The discriminating acceptance check here is: Forecasts retain store IDs and independent store-level holdout errors are available for inspection.

A reviewer can reproduce a feature contribution for a selected prediction and the explanation report does not claim causal proof. This is evidence for a different corrective change: Use supported BigQuery ML model-evaluation and explanation functions at the appropriate global or local scope and distinguish contribution from causation. That other change addresses: A customer support team cannot explain why an SQL-trained BigQuery ML model prioritizes a prediction and treats a coefficient printout as a complete local explanation. It does not validate the specific Westhaven Marketplace remediation for: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. The necessary check instead is: Forecasts retain store IDs and independent store-level holdout errors are available for inspection.

A controlled dataset with nulls produces complete training rows and a per-column missingness audit without label leakage. This is evidence for a different corrective change: Define and fit an explicit BigQuery ML missing-value imputation strategy and track the percentage of imputed records. That other change addresses: A product-cost prediction query silently drops rows with missing numeric features, biasing the downstream training sample. It does not validate the specific Westhaven Marketplace remediation for: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. The necessary check instead is: Forecasts retain store IDs and independent store-level holdout errors are available for inspection.

Held-out user-item recommendations rank relevant items more effectively than a non-personalized popularity baseline. This is evidence for a different corrective change: Use a BigQuery ML recommendation approach such as matrix factorization with appropriate user, item, and interaction data. That other change addresses: A media platform seeks implicit user-item recommendations but trains an ordinary numeric sales regression with user IDs as magnitudes. It does not validate the specific Westhaven Marketplace remediation for: A retailer wants separate forecasts for hundreds of stores but inadvertently aggregates all locations into one time series. The necessary check instead is: Forecasts retain store IDs and independent store-level holdout errors are available for inspection.