AIF-C01 Practice Test 35 — 15 Questions

Question 1

A company uses SageMaker Clarify to detect bias in a deployed model. The monitoring must run automatically on a schedule. Which SageMaker feature should they use?

Accepted Answer

SageMaker Model Monitor. SageMaker Model Monitor can be configured to run bias detection jobs on a schedule using Clarify's bias metrics.

Answer

SageMaker Pipelines

Answer

SageMaker Experiments

Answer

SageMaker Data Wrangler

Question 2

A startup is building a semantic search system over their product catalog using Amazon Bedrock. They want to convert product descriptions into vector embeddings and store them in a vector database for similarity search. Which TWO actions should they take? (Select TWO.)

Accepted Answer

Store the embeddings in a vector database such as Amazon OpenSearch Serverless with the k-NN plugin. Option E is correct because Amazon Titan Embeddings (e.g., amazon.titan-embed-text-v1/v2) is a purpose-built Bedrock text embedding model that converts product descriptions into dense vector embeddings suitable for semantic similarity search. Option A is correct because Amazon OpenSearch Serverless with the k-NN plugin is a managed vector database that supports storing embeddings and running approximate nearest-neighbor (ANN) similarity queries, which is exactly what the semantic search system requires. Option B is wrong because Titan Image Generator produces images from prompts, not embeddings, and it would not generate text-description embeddings. Option C is wrong because Anthropic Claude is a text generation/reasoning model on Bedrock and does not expose an embeddings API via InvokeModel. Option D is wrong because DynamoDB is a key-value/document store without native vector similarity search, so storing vectors as binary attributes would not support k-NN retrieval.

Answer

Use Amazon Titan Image Generator to create embeddings for each product image

Answer

Use Amazon Bedrock InvokeModel with Anthropic Claude to generate embeddings

Answer

Store the embedding vectors in an Amazon DynamoDB table as binary attributes

Question 3

A team is building a classifier to detect fraudulent transactions. The dataset has 99.9% legitimate transactions and 0.1% fraudulent. Which evaluation metric is most appropriate?

Accepted Answer

AUC-ROC. AUC-ROC is the most appropriate metric for this highly imbalanced classification problem because it evaluates the model's ability to distinguish between fraudulent (positive) and legitimate (negative) transactions across all classification thresholds, without being biased by the overwhelming majority of legitimate transactions. Unlike accuracy, AUC-ROC focuses on the trade-off between true positive rate and false positive rate, making it robust to class imbalance where 99.9% of data is negative.

Answer

Mean Absolute Error (MAE)

Answer

Root Mean Squared Error (RMSE)

Answer

Accuracy

Question 4

A data scientist needs to compare the performance of two foundation models on a text summarization task. They have reference summaries for 1000 text samples. Which evaluation approach would provide the MOST reliable comparison?

Accepted Answer

Use both ROUGE scores and human evaluation on a subset. Relying solely on automated metrics like BLEU or ROUGE can be misleading for text summarization, as they measure n-gram overlap but not semantic quality or factual consistency. Combining ROUGE scores (for lexical overlap) with human evaluation on a subset provides a more reliable comparison by capturing both quantitative similarity and qualitative aspects like coherence and faithfulness.

Answer

Use BLEU scores only

Answer

Use ROUGE scores only

Answer

Use human evaluation only

Question 5

A company is building a search application that retrieves relevant documents based on semantic meaning rather than exact keyword matches. Which combination of services would BEST enable this capability?

Accepted Answer

Amazon Titan Embeddings model and a vector database. Amazon Titan Embeddings converts text into dense vector representations that capture semantic meaning, and storing these vectors in a vector database (e.g., Amazon OpenSearch Serverless with vector engine or Aurora PostgreSQL with pgvector) enables similarity search based on cosine distance or Euclidean distance. This combination retrieves documents by semantic proximity rather than exact keyword matches, which is the core requirement for semantic search.

Answer

Amazon Bedrock and Amazon Kendra

Answer

Amazon Lex and Amazon ElastiCache

Answer

Amazon Comprehend and Amazon DynamoDB

Question 6

A company is using Amazon Bedrock to build a code generation assistant for internal developers. They want to reduce costs by processing batch requests during off-peak hours. Which Bedrock feature should they use?

Accepted Answer

Bedrock Batch Inference. B is correct because Amazon Bedrock Batch Inference is specifically designed for processing large volumes of inference requests asynchronously, which allows the company to schedule batch jobs during off-peak hours to reduce costs. This feature leverages lower pricing for asynchronous workloads and is ideal for non-real-time use cases like code generation for internal developers.

Answer

Bedrock Guardrails

Answer

Provisioned Throughput

Answer

Bedrock Agents

Question 7

A financial services company wants to deploy a chatbot using Amazon Bedrock that must never discuss investment advice. Which Bedrock feature should they configure to enforce this policy?

Accepted Answer

Topic denial. Amazon Bedrock Guardrails lets administrators define denied topics (for example, 'investment advice') that the model must not discuss. When a user prompt or model response is detected as related to a denied topic, Guardrails blocks the response and can return a predefined message. This is enforced through Bedrock Guardrails denied topics, not a standalone 'Topic denial' feature.

Answer

Grounding check

Answer

PII detection

Answer

Content filtering

Question 8

A company uses Amazon Bedrock with a Knowledge Base for RAG. Users report that the assistant gives incorrect answers for questions that require understanding of data tables. After reviewing, the team suspects the chunking strategy is breaking table structures. Which change would BEST preserve the integrity of tabular data?

Accepted Answer

Switch from fixed-size chunking to semantic chunking that respects table boundaries. Semantic chunking that respects table boundaries preserves the logical structure of tabular data by ensuring that rows, columns, and headers remain intact within a single chunk. Fixed-size chunking can split a table mid-row or mid-column, causing the knowledge base to retrieve incomplete or misaligned data, which leads to incorrect RAG answers. This approach directly addresses the root cause—broken table structures—without changing the vector store or overlap settings.

Answer

Use a different vector store like Pinecone

Answer

Increase the chunk overlap to 50%

Answer

Decrease the chunk size to 100 tokens

Question 9

A company wants to build a customer service chatbot that answers questions about their internal policy documents. The documents are updated monthly, and the team cannot afford to retrain a model each time. Which approach is MOST appropriate?

Accepted Answer

Use Retrieval-Augmented Generation (RAG) with the policy documents indexed in a vector store. Retrieval-Augmented Generation (RAG) is the most appropriate approach because it allows the chatbot to answer questions based on the latest policy documents without retraining the model. RAG retrieves relevant document chunks from a vector store at inference time and passes them as context to the LLM, ensuring answers are grounded in the most current information. This avoids the cost and complexity of monthly fine-tuning or retraining, while handling the dynamic nature of the documents.

Answer

Fine-tune a base LLM on the policy documents monthly

Answer

Use a larger foundation model with a longer context window and paste all documents into each prompt

Answer

Train a custom model from scratch on the policy documents each month

Question 10

A data scientist wants to compare the text embeddings generated by Amazon Titan Embeddings for a set of product descriptions. Which metric is MOST appropriate to measure the semantic similarity between two embedding vectors?

Accepted Answer

Cosine similarity. Cosine similarity measures the cosine of the angle between two vectors, focusing on their orientation rather than magnitude. For text embeddings from models like Amazon Titan Embeddings, which are normalized to unit length, cosine similarity is the standard metric because it captures semantic similarity even when descriptions differ in length or word count. Euclidean distance would be affected by vector magnitude, making it less suitable for comparing semantic content.

Answer

Hamming distance

Answer

Manhattan distance

Answer

Euclidean distance

Question 11

A company is using Amazon Bedrock to build a chatbot that must comply with data residency regulations. All data must remain in a specific AWS region. Which action is MOST important to meet this requirement?

Question 12

A data scientist is using Amazon Bedrock to build a text summarization application. Which component in Bedrock allows the user to experiment with different models and prompts interactively without writing code?

Accepted Answer

Amazon Bedrock Playground. Amazon Bedrock Playground provides a no-code, interactive interface where users can test and compare different foundation models, adjust prompt parameters, and see real-time responses without writing any code. This directly supports the use case of experimenting with models and prompts for building a text summarization application.

Answer

Amazon Bedrock Guardrails

Answer

Amazon Bedrock Knowledge Bases

Answer

Amazon Bedrock Agents

Question 13

A company is deploying a chatbot using Amazon Bedrock and wants to ensure that the model does not generate offensive or inappropriate content. Which THREE measures can they apply?

Accepted Answer

Use a system prompt to define ethical guidelines and constraints. Option A is correct because a system prompt sets the model's behavioral boundaries and can explicitly instruct it to refuse or avoid offensive, harmful, or inappropriate content, making it a first-line control for responsible generation. Option B is correct because a human-in-the-loop review process catches flagged or borderline responses that automated filters miss, providing an additional safety layer before content reaches users. Option D is correct because Amazon Bedrock Guardrails (and built-in content filters) apply configurable policies such as denied topics, content filters for hate/violence/sexual/insults, and PII redaction to block or mask inappropriate model outputs at inference time. Option C is not appropriate because raising temperature increases randomness and creativity, which generally makes offensive or unpredictable output more likely, not less. Option E is not a reliable standalone measure because fine-tuning on safe conversations does not guarantee the model will never generate offensive content and does not provide runtime enforcement, unlike guardrails or review processes.

Answer

Increase the temperature to make outputs more creative and less likely to repeat offensive phrases

Answer

Fine-tune the model on a dataset of safe conversations only

Question 14

A data scientist needs to convert text into numerical vectors for semantic search. Which type of foundation model should they use?

Accepted Answer

Embedding model. An embedding model is specifically designed to convert text (or other data) into dense numerical vectors that capture semantic meaning. These vectors enable efficient similarity comparisons in vector space, which is the core requirement for semantic search. Text generation models produce sequences of tokens, not fixed-length vector representations, making them unsuitable for this task.

Answer

Text generation model

Answer

Multimodal model

Answer

Image generation model

Question 15

A team is fine-tuning a Meta Llama 2 model on Amazon Bedrock for a legal document classification task. After fine-tuning, the model performs well on the training set but poorly on the validation set. Which adjustment is MOST likely to reduce overfitting?

Accepted Answer

Increase the size of the training dataset and apply dropout. Increasing the training dataset size provides more diverse examples, helping the model generalize better, while dropout randomly deactivates neurons during training to prevent co-adaptation, both of which directly combat overfitting. In the context of fine-tuning Meta Llama 2 on Amazon Bedrock, these techniques are standard regularization methods to improve validation performance when the model memorizes the training set.

Answer

Increase the learning rate

Answer

Add more layers to the model

Answer

Reduce the number of training epochs