Exam Bundle

The training schema references the intended version and a sampled stored record matches its declared transformation lineage. This would verify a different operation: Use explicit versioned feature names and document producer-consumer contracts for the stored feature definitions. That separate action addresses: A SageMaker Feature Store group publishes a transformed feature but training continues to read an older raw column with the same business label. It cannot establish the required Westhaven Retail correction for: A marketing propensity model has a class distribution of 98 percent negative examples and selection by accuracy alone hides poor positive detection. The decisive acceptance check is instead: The final report records positive-class recall and precision on the original distribution, not only inflated aggregate accuracy.

A field-type regression is quarantined and a compatible schema revision remains queryable with the expected typed attributes. This would verify a different operation: Register and validate an explicit schema or compatible evolution policy before promoting events into the curated training zone. That separate action addresses: A newly deployed Glue ingestion job tries to infer the schema of evolving JSON events, and a changed field type silently becomes a null column. It cannot establish the required Westhaven Retail correction for: A marketing propensity model has a class distribution of 98 percent negative examples and selection by accuracy alone hides poor positive detection. The decisive acceptance check is instead: The final report records positive-class recall and precision on the original distribution, not only inflated aggregate accuracy.

An approved source update appears within the freshness objective and replayed change messages do not create duplicate feature rows. This would verify a different operation: Introduce an appropriate CDC or streaming ingestion design that preserves update ordering and deduplicates repeated change events. That separate action addresses: A team uses a daily full-table export for near-real-time ML fraud features even though the source produces frequent small changes. It cannot establish the required Westhaven Retail correction for: A marketing propensity model has a class distribution of 98 percent negative examples and selection by accuracy alone hides poor positive detection. The decisive acceptance check is instead: The final report records positive-class recall and precision on the original distribution, not only inflated aggregate accuracy.

Use an authorized transfer plan with manifest or checksum verification and record the source snapshot identifier for lineage. It does not directly address the Brookside Manufacturing case: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The relevant acceptance criterion for this case is: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

Preserve protected product tokens or use a validated multilingual feature representation that does not corrupt business identifiers. It does not directly address the Brookside Manufacturing case: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The relevant acceptance criterion for this case is: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

Profile the feature distribution by label and apply documented context-aware outlier handling rather than global blind filtering. It does not directly address the Brookside Manufacturing case: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The relevant acceptance criterion for this case is: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

New feature records are queryable from the intended offline store after normal delivery delay while the online store remains functional. This would verify a different operation: Correct the offline-store S3 and IAM configuration and verify feature group creation and data delivery permissions. That separate action addresses: The SageMaker Feature Store offline store is configured with a wrong S3 bucket and the expected historical training features never appear for query. It cannot establish the required Brookside Manufacturing correction for: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The decisive acceptance check is instead: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

A test encrypted object can be read by the approved role while an unrelated role remains denied by S3 or KMS authorization. This would verify a different operation: Grant narrowly scoped source-object and KMS decrypt permissions to the authorized consuming role and confirm the key policy permits the operation. That separate action addresses: An S3 ingestion workload receives intermittent AccessDenied errors because the producing account encrypts objects with a customer-managed KMS key inaccessible to the training role. It cannot establish the required Brookside Manufacturing correction for: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The decisive acceptance check is instead: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

A second training run reads the expected materialized feature version and shows lower repeated transformation time. This would verify a different operation: Materialize governed versioned features with an appropriate offline store and documented recomputation schedule. That separate action addresses: A tabular feature set repeatedly recomputes the same expensive historical aggregations for every training job without a reusable feature definition. It cannot establish the required Brookside Manufacturing correction for: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The decisive acceptance check is instead: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

Use a suitable embedding or hashing approach with collision and memory impact assessed for the model objective. It does not directly address the Bluehaven Transit case: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The relevant acceptance criterion for this case is: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

Preserve protected product tokens or use a validated multilingual feature representation that does not corrupt business identifiers. It does not directly address the Bluehaven Transit case: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The relevant acceptance criterion for this case is: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

Materialize governed versioned features with an appropriate offline store and documented recomputation schedule. It does not directly address the Bluehaven Transit case: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The relevant acceptance criterion for this case is: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

A rerun uses the same immutable S3 manifest and produces identical input checksums without public endpoint dependency. This would verify a different operation: Land a versioned, integrity-checked source copy in an authorized S3 data lake before launching repeatable processing jobs. That separate action addresses: A SageMaker processing job repeatedly downloads a multi-gigabyte model-input dataset from a distant public endpoint and fails on unstable external transfers. It cannot establish the required Bluehaven Transit correction for: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The decisive acceptance check is instead: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

The destination manifest reconciles object counts and checksums with the source snapshot and identifies any failed transfers. This would verify a different operation: Use an authorized transfer plan with manifest or checksum verification and record the source snapshot identifier for lineage. That separate action addresses: A sensitive ML corpus copied by DataSync into S3 has no reproducible record that the transferred objects match the approved source snapshot. It cannot establish the required Bluehaven Transit correction for: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The decisive acceptance check is instead: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

A measured extract contains only authorized columns and the intended time window while transfer volume decreases. This would verify a different operation: Push filters and projections into an appropriate query or unload operation before delivering selected fields to the training pipeline. That separate action addresses: A distributed feature preparation process overfetches every Redshift table column even though the model uses just a few features and one time period. It cannot establish the required Bluehaven Transit correction for: An ML pipeline randomly splits multiple events from one patient across training and test datasets, leaking individual-specific signals. The decisive acceptance check is instead: A cross-split identity audit finds no patient overlap and the held-out metric is computed on previously unseen identities.

Version a single canonical text normalization and tokenizer artifact and use it in both processing and serving containers. It does not directly address the Northgate Telecom case: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The relevant acceptance criterion for this case is: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

Apply an appropriate categorical encoding while preserving the region identifier's nominal semantics and missing-value policy. It does not directly address the Northgate Telecom case: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The relevant acceptance criterion for this case is: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

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 Northgate Telecom case: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The relevant acceptance criterion for this case is: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

A replay of the same source window yields the same per-key feature history despite differing arrival order between partitions. This would verify a different operation: Persist partition and event sequence metadata in the landing dataset and define deterministic replay order for each key. That separate action addresses: A Kinesis-to-S3 feature pipeline creates out-of-order files without documenting stream partition keys, making subsequent replay inconsistent. It cannot establish the required Northgate Telecom correction for: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The decisive acceptance check is instead: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

An unseen category follows the documented fallback and the endpoint receives the same stable feature vector dimensions as training. This would verify a different operation: Fit and persist a categorical encoder on the training set with an explicit unknown-category handling policy for all later inputs. That separate action addresses: A preprocessing notebook one-hot encodes categories separately for training and serving, producing different feature vector dimensions for unseen labels. It cannot establish the required Northgate Telecom correction for: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The decisive acceptance check is instead: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

A field-type regression is quarantined and a compatible schema revision remains queryable with the expected typed attributes. This would verify a different operation: Register and validate an explicit schema or compatible evolution policy before promoting events into the curated training zone. That separate action addresses: A newly deployed Glue ingestion job tries to infer the schema of evolving JSON events, and a changed field type silently becomes a null column. It cannot establish the required Northgate Telecom correction for: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The decisive acceptance check is instead: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

Use distributed Spark transforms or scalable partitioned outputs instead of collecting the full dataset to one process. It does not directly address the Harborline Logistics case: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The relevant acceptance criterion for this case is: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

Materialize governed versioned features with an appropriate offline store and documented recomputation schedule. It does not directly address the Harborline Logistics case: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The relevant acceptance criterion for this case is: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

Use explicit versioned feature names and document producer-consumer contracts for the stored feature definitions. It does not directly address the Harborline Logistics case: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The relevant acceptance criterion for this case is: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

A manifest audit finds every record in exactly its intended split with no overlapping S3 object references. This would verify a different operation: Separate and explicitly version S3 prefixes or manifests for train, validation and test data with role-based write control. That separate action addresses: An S3 training inventory uses object prefixes that mix production and evaluation records, causing the preparation job to include the wrong dataset split. It cannot establish the required Harborline Logistics correction for: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The decisive acceptance check is instead: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

A test encrypted object can be read by the approved role while an unrelated role remains denied by S3 or KMS authorization. This would verify a different operation: Grant narrowly scoped source-object and KMS decrypt permissions to the authorized consuming role and confirm the key policy permits the operation. That separate action addresses: An S3 ingestion workload receives intermittent AccessDenied errors because the producing account encrypts objects with a customer-managed KMS key inaccessible to the training role. It cannot establish the required Harborline Logistics correction for: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The decisive acceptance check is instead: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

A test replay with multiple versions yields one expected customer record and preserves the authorized last-update fields. This would verify a different operation: Deduplicate on stable business identity and event-version rules, retaining the latest accepted entity state as required. That separate action addresses: An ETL pipeline drops duplicate customers using every column as a key even though repeated events have different update timestamps. It cannot establish the required Harborline Logistics correction for: A predictive maintenance model trains on sensor readings collected after a recorded machine failure and reports unrealistic validation accuracy. The decisive acceptance check is instead: A lineage audit confirms all feature timestamps precede the corresponding label decision time and accuracy is recalculated.

Separate and explicitly version S3 prefixes or manifests for train, validation and test data with role-based write control. It does not directly address the Silvercrest University case: A sensitive training corpus contains raw payment card fields that are not needed for the predictive task but are present in shared notebook extracts. The relevant acceptance criterion for this case is: A generated training sample contains no prohibited payment fields and access to the original source is limited to authorized roles.

Profile the feature distribution by label and apply documented context-aware outlier handling rather than global blind filtering. It does not directly address the Silvercrest University case: A sensitive training corpus contains raw payment card fields that are not needed for the predictive task but are present in shared notebook extracts. The relevant acceptance criterion for this case is: A generated training sample contains no prohibited payment fields and access to the original source is limited to authorized roles.

Record and deploy a specific reviewed recipe version together with input snapshot identifiers for every training dataset build. It does not directly address the Silvercrest University case: A sensitive training corpus contains raw payment card fields that are not needed for the predictive task but are present in shared notebook extracts. The relevant acceptance criterion for this case is: A generated training sample contains no prohibited payment fields and access to the original source is limited to authorized roles.

The large synthetic partition completes without driver out-of-memory errors and its aggregate counts match the reference run. This would verify a different operation: Use distributed Spark transforms or scalable partitioned outputs instead of collecting the full dataset to one process. That separate action addresses: A Spark feature preparation stage converts all encoded records to driver memory and fails for large training partitions. It cannot establish the required Silvercrest University correction for: A sensitive training corpus contains raw payment card fields that are not needed for the predictive task but are present in shared notebook extracts. The decisive acceptance check is instead: A generated training sample contains no prohibited payment fields and access to the original source is limited to authorized roles.

A benchmark reads fewer files, preserves event totals and lowers processing overhead for the same approved date partition. This would verify a different operation: Introduce a compaction process that creates suitably sized columnar objects and retains partition and source lineage in the curated dataset. That separate action addresses: An ML data lake receives tiny telemetry objects every second and Spark processing time is dominated by listing and opening millions of files. It cannot establish the required Silvercrest University correction for: A sensitive training corpus contains raw payment card fields that are not needed for the predictive task but are present in shared notebook extracts. The decisive acceptance check is instead: A generated training sample contains no prohibited payment fields and access to the original source is limited to authorized roles.

A sampled reconciliation shows expected one-to-many relationships and the final training row count meets the documented bounds. This would verify a different operation: Define an authoritative join key and a validated join cardinality before merging the two approved data sources. That separate action addresses: An ML team combines order data from RDS and session data from S3 without a stable customer join key and accidentally multiplies training rows. It cannot establish the required Silvercrest University correction for: A sensitive training corpus contains raw payment card fields that are not needed for the predictive task but are present in shared notebook extracts. The decisive acceptance check is instead: A generated training sample contains no prohibited payment fields and access to the original source is limited to authorized roles.