Amazon Web Services MLA-C01 - AWS Certified Machine Learning Engineer - Associate

AWS recommends SageMaker shadow testing as the safest way to validate a new model version using real production traffic without impacting production responses. A shadow variant receives a copy of inference requests but does not return predictions to end users. This completely eliminates the risk of exposing incorrect predictions.

Options A and B are canary deployments. While useful, they still allow v2 to return responses to real users, which violates the requirement to prevent disruption. Option C sends 100% of traffic to v2 externally and is risky.

Shadow variants are explicitly designed to minimize risk while using real data, making them the AWS best practice for pre-production validation.

Therefore, Option D is correct.

Step 1: Create a new model definition with updated settings by using the CreateModel action in the SageMaker AI API.

Step 2: Create a new endpoint configuration that uses the new model definition.

Step 3: Update the endpoint with the new endpoint configuration.

Do not delete and recreate the endpoint. That causes unnecessary inference interruption. SageMaker endpoint updates are designed to use a new EndpointConfig; AWS explicitly states that to update an endpoint, you must create a new endpoint configuration, then call UpdateEndpoint on the existing endpoint. During the update, SageMaker changes the endpoint to Updating and then back to InService.

Option C is correct because the requirement has two separate parts: first, the company must track model versions; second, it needs recommendations for which instance types to use for hosting the model in SageMaker AI. AWS documentation identifies the SageMaker Model Registry as the SageMaker feature used to register and manage model packages and versions as part of the ML lifecycle. For hosting recommendations, AWS documentation says Amazon SageMaker Inference Recommender helps select the best instance type and configuration for ML models and workloads by automating benchmarking and load testing across SageMaker AI instances.

The AWS Inference Recommender documentation is especially important here because it explicitly says you can use it after you register a model to the SageMaker Model Registry with model artifacts. It also states that Inference Recommender helps choose the best endpoint type and configuration and returns recommendations that include the instance type in the resulting endpoint configuration. That is an exact match to the question’s requirement to recommend hosting instance types for a model that has not yet been hosted on SageMaker AI.

The other options do not match AWS service purposes. Amazon ECR stores container images, not model versions. AWS Compute Optimizer is not the SageMaker-native service documented for inference benchmarking of ML models. SageMaker Autopilot is for automated model building, not inference instance recommendations. SageMaker Experiments tracks experiment metadata, trials, and runs, but it does not recommend deployment instance types. Therefore, the fully verified AWS-docs answer is C.

Monitoring bias drift in deployed machine learning models is crucial to ensure fairness and accuracy over time. Amazon SageMaker Clarify provides tools to detect bias in ML models, both during training and after deployment. To monitor bias drift for models deployed to real-time endpoints, an effective approach involves orchestrating SageMaker Clarify jobs using AWS Lambda functions.

Implementation Steps:

Set Up Data Capture:

Enable data capture on the SageMaker endpoint to record input data and model predictions. This captured data serves as the basis for bias analysis.

Develop a Lambda Function:

Create an AWS Lambda function configured to initiate a SageMaker Clarify job. This function will process the captured data to assess bias metrics.

Schedule or Trigger the Lambda Function:

Configure the Lambda function to run on-demand or at scheduled intervals using Amazon CloudWatch Events or EventBridge. This setup allows for regular bias monitoring as per the application ' s requirements.

Analyze and Respond to Results:

After each Clarify job completes, review the generated bias reports. If bias drift is detected, take appropriate actions, such as retraining the model or adjusting data preprocessing steps.

Advantages of This Approach:

Automation: Utilizing AWS Lambda for orchestrating Clarify jobs enables automated and scalable bias monitoring without manual intervention.

Cost-Effectiveness: AWS Lambda ' s serverless nature ensures that you only pay for the compute time consumed during the execution of the function, optimizing resource usage.

Flexibility: The solution can be tailored to specific monitoring needs, allowing for adjustments in monitoring frequency and analysis parameters.

By implementing this solution, the company can effectively monitor bias drift in real-time, ensuring that the AI application maintains fairness and accuracy throughout its lifecycle.

Bias detection is a critical step in responsible machine learning and is emphasized in AWS documentation, particularly in Amazon SageMaker Clarify. When analyzing structured datasets that include sensitive or influential attributes such as age, AWS recommends evaluating label distribution fairness and group-based outcome differences before training a model.

The class imbalance metric helps identify whether certain outcomes (for example, purchase vs. no purchase) are overrepresented or underrepresented. Severe imbalance can cause models to favor majority classes, leading to biased predictions. AWS explicitly highlights class imbalance as a key issue to assess during data exploration.

The Difference in Proportions of Labels (DPL) is a fairness metric supported by SageMaker Clarify that measures whether outcome labels are disproportionately distributed across different groups, such as age ranges. DPL compares the proportion of favorable outcomes between groups, making it especially effective for identifying demographic bias in categorical labels.

Options A and D focus on descriptive statistics or correlations, which are useful for data understanding but do not directly measure bias or fairness. Option B partially addresses imbalance and sentiment but sentiment analysis of reviews alone does not quantify demographic bias tied to outcomes.

AWS documentation strongly recommends using group fairness metrics, including DPL, alongside class imbalance checks to identify bias before training. These metrics provide actionable insights into whether a dataset may lead to unfair or skewed model behavior.

Therefore, Option C is the most appropriate and AWS-aligned choice.

This use case involves large payloads (high-resolution images), variable processing time, and a requirement to notify users upon completion. AWS documentation recommends Amazon SageMaker asynchronous inference for exactly this scenario.

Asynchronous inference allows clients to submit inference requests without waiting for a response. The request payload is stored in Amazon S3, and the inference result is written back to S3 when processing is complete. This approach is designed for workloads that cannot meet real-time latency requirements and where payload sizes exceed real-time limits.

SageMaker asynchronous inference integrates natively with Amazon SNS to send notifications when inference completes or fails, which directly satisfies the user notification requirement.

Batch transform jobs are intended for offline, scheduled processing of large datasets and do not provide built-in user notification mechanisms. Amazon EFS is not required because SageMaker natively integrates with S3 for inference input and output.

AWS explicitly documents asynchronous inference as the preferred solution for large image inference with completion notifications.

Therefore, Option B is the correct and AWS-verified answer.

To minimize communication overhead during distributed training:

1. Same VPC Subnet: Ensures low-latency communication between training instances by keeping the network traffic within a single subnet.

2. Same AWS Region and Availability Zone: Reduces network latency further because cross-AZ communication incurs additional latency and costs.

3. Data in the Same Region and AZ: Ensures that the training data is accessed with minimal latency, improving performance during training.

This configuration optimizes communication efficiency and minimizes overhead.

In a Retrieval Augmented Generation (RAG) architecture, retrieval quality directly impacts response accuracy. AWS documentation for Bedrock and OpenSearch highlights the use of reranker models to improve relevance after initial vector search retrieval.

Vector search retrieves documents based on embedding similarity, but the top results are not always the most contextually relevant. A reranker model evaluates the retrieved documents against the user query and reorders them based on semantic relevance before sending them to the foundation model.

Option A improves readability but does not improve retrieval relevance. Option C filters data before retrieval, which can reduce recall. Option D improves performance, not relevance.

AWS explicitly recommends reranking as a best practice for improving answer quality in RAG systems.

Therefore, Option B is the correct solution.

The primary requirement is dimensionality reduction on high-dimensional structured data to uncover hidden patterns. Principal Component Analysis (PCA) is a linear dimensionality reduction technique specifically designed for this purpose and is available as a built-in algorithm in Amazon SageMaker.

PCA transforms the original features into a smaller set of orthogonal components that preserve the maximum possible variance. This makes PCA ideal for large tabular datasets such as census data, where hundreds of correlated variables are common.

t-SNE (Option B) is mainly used for visualization in very low dimensions (2D or 3D) and does not scale well for large datasets or production analysis. k-means (Option C) is a clustering algorithm, not a dimensionality reduction method. Random Cut Forest (Option D) is used for anomaly detection.

Therefore, PCA is the correct and AWS-recommended solution.