Exam Bundle

A controlled deployment must pass endpoint health, representative inference and alarm checks to reach the completed state. This would verify a different operation: Add post-deployment smoke tests and monitored acceptance gates before the production release is marked successful. That separate action addresses: A team declares its pipeline finished when model training succeeds, though deployment smoke tests and runtime metrics are never observed. It cannot establish the required Maplewood Schools correction for: An auto-scaled inference endpoint scales out during every brief request spike and incurs unnecessary instance churn and warm-up overhead. The decisive acceptance check is instead: A replay of transient spikes avoids oscillation while sustained load still triggers the required scale-out action.

An oversized test input is queued, processed and retrieved asynchronously without the client's synchronous timeout. This would verify a different operation: Choose SageMaker asynchronous inference or a supported job-based flow with input-output object locations and completion handling. That separate action addresses: An image analysis API accepts infrequent large payloads with long processing times but times out when called synchronously through a normal endpoint. It cannot establish the required Maplewood Schools correction for: An auto-scaled inference endpoint scales out during every brief request spike and incurs unnecessary instance churn and warm-up overhead. The decisive acceptance check is instead: A replay of transient spikes avoids oscillation while sustained load still triggers the required scale-out action.

A test build can publish only its approved artifact and is denied direct modification of the production endpoint. This would verify a different operation: Separate pipeline, build, training and serving roles with narrowly scoped permissions for their specific stages. That separate action addresses: A model project uses one broad IAM role for CodePipeline, training and production inference, enabling build jobs to alter runtime resources. It cannot establish the required Maplewood Schools correction for: An auto-scaled inference endpoint scales out during every brief request spike and incurs unnecessary instance churn and warm-up overhead. The decisive acceptance check is instead: A replay of transient spikes avoids oscillation while sustained load still triggers the required scale-out action.

Add an integration test that starts the intended runtime image with the frozen artifact and invokes a representative inference sample. It does not directly address the Lakeside Research case: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The relevant acceptance criterion for this case is: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

Implement a controlled canary or blue-green rollout with CloudWatch alarm gates and automatic rollback to the approved prior version. It does not directly address the Lakeside Research case: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The relevant acceptance criterion for this case is: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

Deploy an appropriately scaled low-latency real-time inference endpoint and route transaction requests through the authorized path. It does not directly address the Lakeside Research case: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The relevant acceptance criterion for this case is: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

A deliberately underperforming candidate remains unapproved and a passing candidate records its evaluation evidence. This would verify a different operation: Add evaluation and conditional model-quality steps linked to the model registry approval process. That separate action addresses: A SageMaker Pipelines workflow trains and deploys without recording evaluation output, so every candidate is automatically approved. It cannot establish the required Lakeside Research correction for: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The decisive acceptance check is instead: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

A representative maximum-size request completes under the chosen platform's supported limits without truncating the response. This would verify a different operation: Compare Lambda, SageMaker asynchronous inference or container hosting based on payload size, memory and duration constraints. That separate action addresses: An inference service needs quick per-document processing with substantial response payloads but a proposed Lambda implementation exceeds its documented resource or duration limits. It cannot establish the required Lakeside Research correction for: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The decisive acceptance check is instead: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

A test candidate with a critical slice regression is blocked despite acceptable aggregate F1 performance. This would verify a different operation: Include explicitly approved per-segment acceptance checks and fairness review in the conditional deployment gates. That separate action addresses: A quality gate evaluates only global F1 score while a new candidate violates the acceptable error limit for a protected segment. It cannot establish the required Lakeside Research correction for: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The decisive acceptance check is instead: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

Measure cache behavior and choose suitable memory capacity or a dedicated deployment for the critical high-frequency model. It does not directly address the Eastshore Energy case: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The relevant acceptance criterion for this case is: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

Choose a private supported endpoint and VPC integration design with restricted consumers and approved ingress controls. It does not directly address the Eastshore Energy case: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The relevant acceptance criterion for this case is: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

Benchmark end-to-end inference latency and select an approved Region or edge design aligned to residency and client location. It does not directly address the Eastshore Energy case: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The relevant acceptance criterion for this case is: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

A staged rollback restores the approved prior model version and representative requests return to baseline behavior. This would verify a different operation: Maintain versioned model artifacts and use a controlled deployment strategy with retained previous endpoint configuration. That separate action addresses: A production team deploys an updated fraud model by replacing the only model artifact, leaving no usable rollback option. It cannot establish the required Eastshore Energy correction for: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The decisive acceptance check is instead: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

The endpoint becomes healthy and returns a valid prediction for a documented JSON request through the SageMaker invocation API. This would verify a different operation: Implement the required model loading, health and invocation handlers for the supported SageMaker custom container interface. That separate action addresses: A platform team deploys a custom PyTorch inference container without the correct SageMaker-compatible serving endpoint contract. It cannot establish the required Eastshore Energy correction for: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The decisive acceptance check is instead: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

A known request traverses the versioned preprocessing and prediction stages and produces the expected output after deployment. This would verify a different operation: Use a supported SageMaker inference pipeline or reviewed multi-container design with compatible input and output contracts. That separate action addresses: An inference service chains preprocessing and prediction in application code that cannot be versioned with the SageMaker deployed model. It cannot establish the required Eastshore Energy correction for: A data scientist provisions GPUs for all endpoint variants even though most deployed lightweight models support cost-effective CPU hosting. The decisive acceptance check is instead: The updated resource inventory shows justified instance families and the service maintains required per-variant latency.

Make the retraining input a named immutable dataset manifest and persist it in the pipeline execution parameters. It does not directly address the Riverton Utilities case: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The relevant acceptance criterion for this case is: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

Filter EventBridge or S3-trigger inputs to the intended versioned training dataset and apply an idempotent retraining key. It does not directly address the Riverton Utilities case: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The relevant acceptance criterion for this case is: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

Compare supported CPU, GPU or inference-optimized instances using matched representative throughput and latency tests. It does not directly address the Riverton Utilities case: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The relevant acceptance criterion for this case is: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

A replay of the execution reads identical object versions and generates a traceable, comparable training run. This would verify a different operation: Make the retraining input a named immutable dataset manifest and persist it in the pipeline execution parameters. That separate action addresses: A scheduled retraining pipeline uses the latest available dataset rather than the dataset cut recorded in its review ticket. It cannot establish the required Riverton Utilities correction for: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The decisive acceptance check is instead: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

An unapproved model package is rejected and only a reviewed approved package reaches the endpoint update step. This would verify a different operation: Require the intended registered model package approval state and enforce a deployment role boundary at the promotion stage. That separate action addresses: A CodePipeline deployment stage accepts any model registry package regardless of its manual approval status. It cannot establish the required Riverton Utilities correction for: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The decisive acceptance check is instead: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

The representative load test meets latency targets and the chosen hosting cost falls below the idle GPU baseline. This would verify a different operation: Benchmark CPU or suitable serverless hosting and select the deployment offering that meets utilization and latency requirements. That separate action addresses: A team serves a tiny low-traffic model on a provisioned GPU endpoint even though CPU inference meets the request latency target. It cannot establish the required Riverton Utilities correction for: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The decisive acceptance check is instead: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

Use unique version identifiers, idempotent promotion operations and recoverable state tracking around the deployment stage. It does not directly address the Cedarfield Hospitality case: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The relevant acceptance criterion for this case is: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

Deploy an appropriately scaled low-latency real-time inference endpoint and route transaction requests through the authorized path. It does not directly address the Cedarfield Hospitality case: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The relevant acceptance criterion for this case is: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

Implement a controlled canary or blue-green rollout with CloudWatch alarm gates and automatic rollback to the approved prior version. It does not directly address the Cedarfield Hospitality case: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The relevant acceptance criterion for this case is: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

A synthetic peak test reaches the latency objective while instance count scales back after demand subsides. This would verify a different operation: Choose autoscaled real-time hosting with appropriate minimum capacity and measured scaling response to the traffic pattern. That separate action addresses: An inference workload receives highly variable request bursts and a fixed number of endpoint instances causes long queues at peak. It cannot establish the required Cedarfield Hospitality correction for: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The decisive acceptance check is instead: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

The deployed task reports the exact image digest emitted by the reviewed build stage. This would verify a different operation: Pass an immutable ECR image digest as a build artifact to the deployment stage rather than reconstructing a tag. That separate action addresses: A CodeBuild job pushes an inference image to ECR but the subsequent deploy stage uses a different mutable image tag. It cannot establish the required Cedarfield Hospitality correction for: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The decisive acceptance check is instead: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

Two identical trigger messages resolve to the expected single logical model update without competing untracked releases. This would verify a different operation: Include stable dataset IDs and idempotency controls in the orchestration design, preserving run lineage and concurrency policies. That separate action addresses: A retraining pipeline can be launched twice for the same input data, creating competing model versions with indistinguishable metadata. It cannot establish the required Cedarfield Hospitality correction for: A deployment template grants one IAM role both model-serving permissions and broad stack-administration privileges unrelated to inference. The decisive acceptance check is instead: The runtime can read its designated artifacts and log streams but cannot modify unrelated CloudFormation stacks.

Make the retraining input a named immutable dataset manifest and persist it in the pipeline execution parameters. It does not directly address the Alpine Healthcare case: A container inference service sends health checks to a path that triggers costly full model predictions and exhausts request capacity during rollouts. The relevant acceptance criterion for this case is: Health probes confirm readiness without increasing model invocation counts or delaying normal business predictions.

Design the artifact sharing, replication and KMS policies explicitly for each trusted cross-account stage. It does not directly address the Alpine Healthcare case: A container inference service sends health checks to a path that triggers costly full model predictions and exhausts request capacity during rollouts. The relevant acceptance criterion for this case is: Health probes confirm readiness without increasing model invocation counts or delaying normal business predictions.

Pass an immutable ECR image digest as a build artifact to the deployment stage rather than reconstructing a tag. It does not directly address the Alpine Healthcare case: A container inference service sends health checks to a path that triggers costly full model predictions and exhausts request capacity during rollouts. The relevant acceptance criterion for this case is: Health probes confirm readiness without increasing model invocation counts or delaying normal business predictions.

A representative maximum-size request completes under the chosen platform's supported limits without truncating the response. This would verify a different operation: Compare Lambda, SageMaker asynchronous inference or container hosting based on payload size, memory and duration constraints. That separate action addresses: An inference service needs quick per-document processing with substantial response payloads but a proposed Lambda implementation exceeds its documented resource or duration limits. It cannot establish the required Alpine Healthcare correction for: A container inference service sends health checks to a path that triggers costly full model predictions and exhausts request capacity during rollouts. The decisive acceptance check is instead: Health probes confirm readiness without increasing model invocation counts or delaying normal business predictions.

Representative models load on demand and per-model latency and isolation checks meet the accepted limits at reduced aggregate cost. This would verify a different operation: Evaluate SageMaker multi-model endpoints where model memory and isolation requirements permit compatible shared hosting. That separate action addresses: A production endpoint hosts hundreds of small customer-specific models on separate instances, creating high fixed infrastructure cost. It cannot establish the required Alpine Healthcare correction for: A container inference service sends health checks to a path that triggers costly full model predictions and exhausts request capacity during rollouts. The decisive acceptance check is instead: Health probes confirm readiness without increasing model invocation counts or delaying normal business predictions.

The selected instance family meets throughput and latency targets with a documented cost-per-inference measurement. This would verify a different operation: Compare supported CPU, GPU or inference-optimized instances using matched representative throughput and latency tests. That separate action addresses: A high-volume endpoint runs on small CPUs despite a documented model benchmark showing matrix inference benefits from accelerator hardware. It cannot establish the required Alpine Healthcare correction for: A container inference service sends health checks to a path that triggers costly full model predictions and exhausts request capacity during rollouts. The decisive acceptance check is instead: Health probes confirm readiness without increasing model invocation counts or delaying normal business predictions.