Exam Bundle

A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities. This would verify a different operation: Split examples by patient identity and keep all records for one identity in only one dataset partition. That separate action addresses: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. It cannot establish the required Highland Education correction for: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The decisive acceptance check is instead: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

An audit of transformed images preserves orientation-sensitive labels and prohibited flips are absent from model training. This would verify a different operation: Restrict image augmentation to clinically valid transformations and verify label invariance for each permitted operation. That separate action addresses: A clinical image dataset uses an augmentation pipeline that reverses left-right orientation despite labels depending on laterality. It cannot establish the required Highland Education correction for: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The decisive acceptance check is instead: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

A controlled rebuild with the same recipe and source manifest reproduces feature checksums and row counts. This would verify a different operation: Record and deploy a specific reviewed recipe version together with input snapshot identifiers for every training dataset build. That separate action addresses: An AWS Glue DataBrew recipe is edited directly in production and a rerun cannot reproduce the original training features. It cannot establish the required Highland Education correction for: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The decisive acceptance check is instead: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

Standardize image dimensions, channel ordering and value normalization in the shared preprocessing artifact. It does not directly address the Maplewood Schools case: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The relevant acceptance criterion for this case is: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

Use a stratified or otherwise class-aware splitting method with declared minimum representation checks for the holdout set. It does not directly address the Maplewood Schools case: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The relevant acceptance criterion for this case is: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

Record and deploy a specific reviewed recipe version together with input snapshot identifiers for every training dataset build. It does not directly address the Maplewood Schools case: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The relevant acceptance criterion for this case is: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

Evaluation documents with relevant late sections contribute to the resulting model input and the length limits remain enforced. This would verify a different operation: Adopt a tested document chunking and aggregation strategy aligned to the model's maximum input length. That separate action addresses: An NLP preparation stage truncates every long document to its first paragraph, systematically dropping the labeled information near the end. It cannot establish the required Maplewood Schools correction for: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The decisive acceptance check is instead: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

The duplicate audit shows coherent labels within matching-image clusters and a reviewed exception log tracks ambiguous cases. This would verify a different operation: Fingerprint and review duplicate image groups and reconcile their class labels before publishing the approved training set. That separate action addresses: A classification training set contains duplicated near-identical product images assigned conflicting labels, creating avoidable noise. It cannot establish the required Maplewood Schools correction for: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The decisive acceptance check is instead: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

A controlled event series produces identical seven-day feature values in both training and real-time serving paths. This would verify a different operation: Implement the same window boundaries, event-time cutoff and aggregation logic for offline and online feature creation. That separate action addresses: A training set combines click counts over seven-day windows but the online feature computation uses calendar-week counts instead. It cannot establish the required Maplewood Schools correction for: A data ingestion script assumes S3 is a POSIX filesystem and tries in-place append operations on one object during batch preparation. The decisive acceptance check is instead: Subsequent readers select the newly committed version and no partial object contents are exposed as an accepted input.

Apply retention and lifecycle rules compatible with artifact lineage and archive immutable source objects required for approved reproducibility. It does not directly address the Lakeside Research case: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The relevant acceptance criterion for this case is: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

Classify regional attributes and keep processing and storage in approved AWS Regions with authorized data residency controls. It does not directly address the Lakeside Research case: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The relevant acceptance criterion for this case is: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

Version a single canonical text normalization and tokenizer artifact and use it in both processing and serving containers. It does not directly address the Lakeside Research case: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The relevant acceptance criterion for this case is: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

The baseline report contains relevant protected-group comparisons and an accountable review documents any mitigation and residual risk. This would verify a different operation: Run the appropriate Clarify data-bias analysis and record the fairness baseline and approved mitigation assessment before model release. That separate action addresses: SageMaker Clarify pre-training bias metrics flag a demographic proportion imbalance but the team reports only overall training loss. It cannot establish the required Lakeside Research correction for: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The decisive acceptance check is instead: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

A similarity check finds no synthetic descendants in test data and the external holdout result reflects genuine generalization. This would verify a different operation: Split the original records first, then apply synthetic oversampling exclusively to the training split. That separate action addresses: A fraud model is trained on synthetic minority samples generated before splitting the source data, leading to near-duplicate examples in the test set. It cannot establish the required Lakeside Research correction for: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The decisive acceptance check is instead: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

A historical example contains the category active at its original event timestamp instead of a later corrected label. This would verify a different operation: Use time-aware joins against effective-dated reference values to reconstruct the features available when each event occurred. That separate action addresses: A batch transformation joins slowly changing reference categories using today's labels for all historic examples, leaking future feature state. It cannot establish the required Lakeside Research correction for: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The decisive acceptance check is instead: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

Protect the token mapping with separate least-privilege access and encryption controls and limit its retention to approved workflows. It does not directly address the Eastshore Energy case: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The relevant acceptance criterion for this case is: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

Perform approved irreversible masking or tokenization before feature extraction and restrict the original identifier fields. It does not directly address the Eastshore Energy case: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The relevant acceptance criterion for this case is: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

Standardize image dimensions, channel ordering and value normalization in the shared preprocessing artifact. It does not directly address the Eastshore Energy case: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The relevant acceptance criterion for this case is: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

Model input checks confirm no implied numerical ordering and all expected regions map to the approved encoded features. This would verify a different operation: Apply an appropriate categorical encoding while preserving the region identifier's nominal semantics and missing-value policy. That separate action addresses: A numeric categorical code such as region ID is mistakenly treated as an ordered magnitude when supplied to a linear model. It cannot establish the required Eastshore Energy correction for: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The decisive acceptance check is instead: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

A compliance scan finds sensitive samples only in their permitted regions and a disallowed cross-region copy is denied. This would verify a different operation: Classify regional attributes and keep processing and storage in approved AWS Regions with authorized data residency controls. That separate action addresses: A data preparation flow mixes historical US and EU user records without enforcing the documented regional storage restriction for personal attributes. It cannot establish the required Eastshore Energy correction for: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The decisive acceptance check is instead: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

A deliberately labeled test row never exposes its target through the predictor schema and the suspicious accuracy is re-evaluated. This would verify a different operation: Implement a schema-level separation of label and predictor columns and verify feature lineage prior to fitting the model. That separate action addresses: A data scientist encodes the target label as an input feature for convenience and obtains a nearly perfect training score. It cannot establish the required Eastshore Energy correction for: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. The decisive acceptance check is instead: An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows.

Make the feature write idempotent using event identifiers or version checks and send repeated failures to a controlled recovery path. It does not directly address the Riverton Utilities case: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The relevant acceptance criterion for this case is: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

Fit and persist a categorical encoder on the training set with an explicit unknown-category handling policy for all later inputs. It does not directly address the Riverton Utilities case: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The relevant acceptance criterion for this case is: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

Use chronological training, validation and final holdout windows with a gap where leakage risk requires one. It does not directly address the Riverton Utilities case: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The relevant acceptance criterion for this case is: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

A bilingual reference set retains the required product terms and its label agreement remains inside the approved quality range. This would verify a different operation: Preserve protected product tokens or use a validated multilingual feature representation that does not corrupt business identifiers. That separate action addresses: A sentiment pipeline blindly translates all customer messages into English even though translation changes domain-specific product identifiers. It cannot establish the required Riverton Utilities correction for: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The decisive acceptance check is instead: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

Every required rare class appears at least the configured minimum count in evaluation and no training row is duplicated into test. This would verify a different operation: Use a stratified or otherwise class-aware splitting method with declared minimum representation checks for the holdout set. That separate action addresses: A multi-class label dataset is split without stratification and a rare class is missing entirely from the evaluation partition. It cannot establish the required Riverton Utilities correction for: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The decisive acceptance check is instead: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

The faulty batch is quarantined based on the defined feature threshold and a clean control batch continues processing. This would verify a different operation: Add schema, null-rate and range checks using AWS Glue Data Quality or equivalent before the training step begins. That separate action addresses: An automated data quality gate checks only row counts while a critical feature's null rate jumps from one percent to 40 percent. It cannot establish the required Riverton Utilities correction for: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The decisive acceptance check is instead: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

Deduplicate on stable business identity and event-version rules, retaining the latest accepted entity state as required. It does not directly address the Cedarfield Hospitality case: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The relevant acceptance criterion for this case is: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

Run the appropriate Clarify data-bias analysis and record the fairness baseline and approved mitigation assessment before model release. It does not directly address the Cedarfield Hospitality case: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The relevant acceptance criterion for this case is: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

Add schema, null-rate and range checks using AWS Glue Data Quality or equivalent before the training step begins. It does not directly address the Cedarfield Hospitality case: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The relevant acceptance criterion for this case is: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

Every required rare class appears at least the configured minimum count in evaluation and no training row is duplicated into test. This would verify a different operation: Use a stratified or otherwise class-aware splitting method with declared minimum representation checks for the holdout set. That separate action addresses: A multi-class label dataset is split without stratification and a rare class is missing entirely from the evaluation partition. It cannot establish the required Cedarfield Hospitality correction for: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The decisive acceptance check is instead: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

The feature vector dimensions stay bounded and an evaluation report measures collisions or embedding quality on held-out accounts. This would verify a different operation: Use a suitable embedding or hashing approach with collision and memory impact assessed for the model objective. That separate action addresses: A recommendation model's high-cardinality account IDs explode into impractically wide one-hot sparse features. It cannot establish the required Cedarfield Hospitality correction for: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The decisive acceptance check is instead: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.

The faulty batch is quarantined based on the defined feature threshold and a clean control batch continues processing. This would verify a different operation: Add schema, null-rate and range checks using AWS Glue Data Quality or equivalent before the training step begins. That separate action addresses: An automated data quality gate checks only row counts while a critical feature's null rate jumps from one percent to 40 percent. It cannot establish the required Cedarfield Hospitality correction for: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. The decisive acceptance check is instead: A measured extract contains only authorized columns and the intended time window while transfer volume decreases.