Exam Bundle

The chosen checkpoint stops before sustained validation degradation and improves the untouched test-set performance. This would verify a different operation: Apply an appropriate regularization or early-stopping strategy selected against validation behavior instead of increasing epochs blindly. That separate action addresses: A production classifier's training loss steadily decreases while validation loss rises, indicating the model is memorizing the training examples. It cannot establish the required Keystone BioLabs correction for: A model validation workflow repeatedly tunes its hyperparameters on the same final test set, inflating the reported out-of-sample quality. The decisive acceptance check is instead: The experiment record shows no selection decisions based on final test labels before the locked evaluation.

The feasibility report identifies supported series and validates forecasts without silently fabricating missing chronological observations. This would verify a different operation: Assess temporal data sufficiency, sampling frequency and known-future covariates before comparing time-series models. That separate action addresses: An ML engineer chooses a built-in forecasting method without confirming that multiple product series contain sufficient and regularly spaced history. It cannot establish the required Keystone BioLabs correction for: A model validation workflow repeatedly tunes its hyperparameters on the same final test set, inflating the reported out-of-sample quality. The decisive acceptance check is instead: The experiment record shows no selection decisions based on final test labels before the locked evaluation.

An interruption test restarts from the last saved checkpoint rather than the initial epoch and preserves model state. This would verify a different operation: Enable appropriate SageMaker managed Spot training checkpoints in durable storage and a tested resume workflow. That separate action addresses: A lengthy training job uses Spot Instances without checkpoints and loses all progress when a capacity interruption stops the instance. It cannot establish the required Keystone BioLabs correction for: A model validation workflow repeatedly tunes its hyperparameters on the same final test set, inflating the reported out-of-sample quality. The decisive acceptance check is instead: The experiment record shows no selection decisions based on final test labels before the locked evaluation.

Include interpretability and governance constraints in model selection and compare explainable models or validated explanation techniques. It does not directly address the Granite SaaS case: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The relevant acceptance criterion for this case is: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

Validate tokenizer and embedding compatibility or use the correct documented resize and artifact versioning procedure. It does not directly address the Granite SaaS case: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The relevant acceptance criterion for this case is: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

Define meaningful parameter ranges, metric objectives and SageMaker tuning job resource and stopping limits. It does not directly address the Granite SaaS case: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The relevant acceptance criterion for this case is: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

A clean isolated training job starts without public internet egress and logs the expected dependency versions. This would verify a different operation: Package dependencies in a trusted container image or approved private repository accessible to the training subnet. That separate action addresses: A model training run terminates because it assumes access to external Python package indexes blocked by the organization's isolated VPC. It cannot establish the required Granite SaaS correction for: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The decisive acceptance check is instead: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

A holdout run produces numeric unit forecasts and is evaluated with documented regression or forecasting error metrics. This would verify a different operation: Select an appropriate regression or time-series modeling approach aligned to the continuous demand target and its evaluation horizon. That separate action addresses: An inventory forecasting team proposes a binary classification algorithm although the desired output is a continuous quantity of product units. It cannot establish the required Granite SaaS correction for: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The decisive acceptance check is instead: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

The same container trains successfully with a different authorized input channel and no code edits. This would verify a different operation: Use SageMaker's configured channel path or environment variables for training data and remove developer-specific bucket references. That separate action addresses: A custom SageMaker training script downloads its dataset from a hard-coded developer bucket rather than reading the configured training input channel. It cannot establish the required Granite SaaS correction for: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The decisive acceptance check is instead: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

Inspect training diagnostics and apply suitable learning-rate reduction or gradient clipping with controlled numerical checks. It does not directly address the Westhaven Retail case: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The relevant acceptance criterion for this case is: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

Enable appropriate SageMaker managed Spot training checkpoints in durable storage and a tested resume workflow. It does not directly address the Westhaven Retail case: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The relevant acceptance criterion for this case is: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

Choose a calibrated regression or probabilistic forecasting method compatible with the numeric target and uncertainty reporting requirement. It does not directly address the Westhaven Retail case: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The relevant acceptance criterion for this case is: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

A completed job records a retrievable model artifact and a clean inference container loads the expected weights. This would verify a different operation: Write trained artifacts to the supported SageMaker model artifact path and confirm the training job uploads the final tarball to its configured S3 output. That separate action addresses: A SageMaker PyTorch training job saves its learned weights only to ephemeral container memory and cannot produce a deployable model artifact on completion. It cannot establish the required Westhaven Retail correction for: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The decisive acceptance check is instead: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

The selected model meets the documented predictive threshold and produces auditable feature-level explanations for controlled cases. This would verify a different operation: Include interpretability and governance constraints in model selection and compare explainable models or validated explanation techniques. That separate action addresses: A fraud classifier must explain individual high-impact denials, yet the team chooses an opaque architecture solely for its slightly higher validation accuracy. It cannot establish the required Westhaven Retail correction for: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The decisive acceptance check is instead: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

A reference document can receive multiple correct labels and per-label quality is assessed on a fixed holdout. This would verify a different operation: Choose a multi-label output design with independent label scores and threshold selection instead of single-class competition. That separate action addresses: A business team asks a multi-label model to assign several valid tags per document but data scientists configure mutually exclusive softmax classes. It cannot establish the required Westhaven Retail correction for: A production candidate has excellent mean latency but fails the request SLA during high-percentile load spikes. The decisive acceptance check is instead: The release report includes tail latency under load and the candidate meets the agreed high-percentile objective.

Evaluate a sequence-aware recommendation or ranking approach and compare it with a simpler nonsequential baseline. It does not directly address the Brookside Manufacturing case: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The relevant acceptance criterion for this case is: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

Introduce validated label-preserving augmentation on training data and retain the unchanged holdout image set. It does not directly address the Brookside Manufacturing case: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The relevant acceptance criterion for this case is: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

Configure rank-aware data sharding or distributed sampling so every worker receives the intended subset for the epoch. It does not directly address the Brookside Manufacturing case: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The relevant acceptance criterion for this case is: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

The selection report compares test metrics, training cost and operational complexity on a common fixed dataset. This would verify a different operation: Benchmark suitable tabular algorithms such as SageMaker XGBoost against the complex model using the same split and budget. That separate action addresses: An analyst trains a high-capacity neural network on a small, mostly numeric transaction table without first benchmarking simpler tabular methods. It cannot establish the required Brookside Manufacturing correction for: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The decisive acceptance check is instead: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

The same container trains successfully with a different authorized input channel and no code edits. This would verify a different operation: Use SageMaker's configured channel path or environment variables for training data and remove developer-specific bucket references. That separate action addresses: A custom SageMaker training script downloads its dataset from a hard-coded developer bucket rather than reading the configured training input channel. It cannot establish the required Brookside Manufacturing correction for: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The decisive acceptance check is instead: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

A controlled repeat uses the same artifact hashes and produces model metrics inside its declared reproducibility tolerance. This would verify a different operation: Pin the source revision, container digest, package versions, seeds and data manifest in the experiment configuration. That separate action addresses: A SageMaker training step is rerun months later and produces a different outcome because the container retrieves unpinned dependency versions. It cannot establish the required Brookside Manufacturing correction for: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The decisive acceptance check is instead: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

Isolate each experiment run's checkpoints, model outputs and parameter metadata under stable unique identifiers. It does not directly address the Bluehaven Transit case: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The relevant acceptance criterion for this case is: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

Introduce validated label-preserving augmentation on training data and retain the unchanged holdout image set. It does not directly address the Bluehaven Transit case: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The relevant acceptance criterion for this case is: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

Evaluate SageMaker-supported distributed data parallel training with synchronized optimizer state and consistent data sharding. It does not directly address the Bluehaven Transit case: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The relevant acceptance criterion for this case is: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

The tuning experiment completes inside its resource cap and produces an auditable best-candidate comparison. This would verify a different operation: Define meaningful parameter ranges, metric objectives and SageMaker tuning job resource and stopping limits. That separate action addresses: An ML engineer runs random search without bounding costly hyperparameters and repeatedly launches training jobs beyond the approved experiment budget. It cannot establish the required Bluehaven Transit correction for: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The decisive acceptance check is instead: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

The model trains without GPU out-of-memory errors and produces comparable validation metrics under the selected memory budget. This would verify a different operation: Reduce the effective batch or apply supported gradient accumulation and mixed precision after checking numerical stability. That separate action addresses: A high-memory feature matrix repeatedly exhausts GPU memory because training uses an unnecessarily large batch size. It cannot establish the required Bluehaven Transit correction for: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The decisive acceptance check is instead: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

A repeated training run keeps finite gradients and loss while meeting the declared convergence criterion. This would verify a different operation: Inspect training diagnostics and apply suitable learning-rate reduction or gradient clipping with controlled numerical checks. That separate action addresses: A custom model has an unstable optimizer and loss becomes nonfinite after a few updates due to excessive gradient magnitudes. It cannot establish the required Bluehaven Transit correction for: A regression model produces implausible forecasts for extreme inputs but the team inspects only the overall RMSE. The decisive acceptance check is instead: The slice report exposes extreme-input errors and documents whether the corrective model stays within accepted thresholds.

Evaluate an unsupervised clustering approach and select a suitable distance representation before proposing any business segment labels. It does not directly address the Northgate Telecom case: SageMaker Clarify generates feature attribution values but the report is interpreted as proof that those features causally determine the outcome. The relevant acceptance criterion for this case is: The governance report distinguishes predictive contribution from causation and records the limits of SHAP-style explanations.

Benchmark a robust loss or documented outlier-aware training strategy using an appropriate validation error criterion. It does not directly address the Northgate Telecom case: SageMaker Clarify generates feature attribution values but the report is interpreted as proof that those features causally determine the outcome. The relevant acceptance criterion for this case is: The governance report distinguishes predictive contribution from causation and records the limits of SHAP-style explanations.

Apply an appropriate regularization or early-stopping strategy selected against validation behavior instead of increasing epochs blindly. It does not directly address the Northgate Telecom case: SageMaker Clarify generates feature attribution values but the report is interpreted as proof that those features causally determine the outcome. The relevant acceptance criterion for this case is: The governance report distinguishes predictive contribution from causation and records the limits of SHAP-style explanations.

The compressed model meets the artifact-size target and its measured accuracy loss stays within the approved limit. This would verify a different operation: Evaluate supported pruning or quantization techniques with quality and compatibility checks for the intended runtime. That separate action addresses: A model developer tries to reduce a large inference artifact by removing random trained layers without validation or a deployment-size target. It cannot establish the required Northgate Telecom correction for: SageMaker Clarify generates feature attribution values but the report is interpreted as proof that those features causally determine the outcome. The decisive acceptance check is instead: The governance report distinguishes predictive contribution from causation and records the limits of SHAP-style explanations.

The candidate evaluation demonstrates compliant request routing and authorized data handling as well as acceptable answer quality. This would verify a different operation: Compare model and deployment options against data residency, access policy and cost constraints before deciding on Bedrock or a supported hosted alternative. That separate action addresses: An inference team chooses a managed language model with the largest context window even though the protected request content cannot leave its designated processing environment. It cannot establish the required Northgate Telecom correction for: SageMaker Clarify generates feature attribution values but the report is interpreted as proof that those features causally determine the outcome. The decisive acceptance check is instead: The governance report distinguishes predictive contribution from causation and records the limits of SHAP-style explanations.

A holdout run produces numeric unit forecasts and is evaluated with documented regression or forecasting error metrics. This would verify a different operation: Select an appropriate regression or time-series modeling approach aligned to the continuous demand target and its evaluation horizon. That separate action addresses: An inventory forecasting team proposes a binary classification algorithm although the desired output is a continuous quantity of product units. It cannot establish the required Northgate Telecom correction for: SageMaker Clarify generates feature attribution values but the report is interpreted as proof that those features causally determine the outcome. The decisive acceptance check is instead: The governance report distinguishes predictive contribution from causation and records the limits of SHAP-style explanations.