Amazon AWS Certified Generative AI Developer - Professional AIP-C01 Free Practice Test - 20 Questions

Question 1

A developer is building a chatbot using Amazon Bedrock. The chatbot needs to answer questions based on a large internal company document that is updated weekly. The developer has only a small set of labeled example conversations. Which approach should the developer use to minimize cost and development effort while ensuring the model can reference the latest document content?

Accepted Answer

Use Retrieval-Augmented Generation (RAG) with a vector database to retrieve relevant document chunks at inference time.

Question 2

Which of the following is a key consideration when selecting a foundation model in Amazon Bedrock for a real-time customer service chatbot?

Accepted Answer

Inference latency requirements

Question 3

A company needs to generate personalized marketing images for thousands of customers daily. The images are created by invoking an Amazon Bedrock model for each customer. The workload runs for 8 hours each night and is time-sensitive. Which inference option should the company choose to minimize cost while meeting the performance requirements?

Accepted Answer

On-Demand Throughput without any commitment

Question 4

A developer receives a 'ModelNotFoundException' when invoking a foundation model in Amazon Bedrock. What is the most likely cause?

Accepted Answer

The model ID specified in the request is incorrect or has been deprecated.

Question 5

Which AWS feature allows developers to enforce safety policies on model outputs to filter harmful content?

Accepted Answer

Amazon Bedrock Guardrails

Question 6

A developer is building a Q&A application over a large technical documentation set using Amazon Bedrock. The documents are stored as PDFs in Amazon S3. The solution must provide answers from the most recent version of the documents with low latency. Which approach should the developer use?

Accepted Answer

Create an Amazon Bedrock Knowledge Base with S3 as the data source, and configure chunking and vector storage.

Question 7

A company is using an Amazon Bedrock foundation model on-demand for a text summarization application. The cost is higher than expected. Analysis shows that many requests include extremely long input documents. Which action is most effective in reducing cost?

Accepted Answer

Implement input token trimming and limit the prompt length to the minimum necessary.

Question 8

A developer needs to ensure that network traffic between a VPC and Amazon Bedrock does not traverse the public internet. Which configuration should be used?

Accepted Answer

Create a VPC endpoint for Bedrock using AWS PrivateLink.

Question 9

A developer wants to adapt a large language model on a small dataset of company-specific examples without losing the model's general knowledge. Which customization technique is most appropriate?

Accepted Answer

Parameter-efficient fine-tuning (PEFT) such as LoRA

Question 10

A developer is building a real-time chat application using Amazon Bedrock. The application must provide fast retrieval of relevant information from a knowledge base. Which vector store configuration should be used to minimize latency?

Accepted Answer

Use Amazon OpenSearch Serverless with provisioned indexing capacity.

Question 11

A Lambda function that invokes Amazon Bedrock occasionally fails with 'RuntimeException' and the error message indicates 'SDK client timeout'. What is the most likely cause?

Accepted Answer

The Lambda function's timeout setting is too short for the model invocation to complete.

Question 12

A company is using multiple versions of a foundation model in Amazon Bedrock for testing. They want to ensure that a production application always uses the latest approved version without manual code updates. What should be done?

Accepted Answer

Use an Amazon Bedrock model alias and point it to the latest version, then update the alias when a new version is approved.

Question 13

A news media company wants to deploy a generative AI application that must have extremely low latency for readers across Europe and Asia. The application uses Amazon Bedrock with a large language model. Which architecture best meets the latency requirement?

Accepted Answer

Deploy the application in multiple AWS regions and use Amazon Route 53 latency-based routing to direct traffic to the nearest region.

Question 14

A developer is evaluating the cost of an Amazon Bedrock application that processes user queries. The application invokes a foundation model multiple times per query to perform reasoning. Which pricing component is most critical to estimate the total cost?

Accepted Answer

Number of input and output tokens per inference multiplied by the number of inferences per query

Question 15

A data science team wants to evaluate the performance of a custom fine-tuned model on a specific task. They have a labeled test dataset. Which tool can be used to run standardized evaluations?

Accepted Answer

Amazon Bedrock Model Evaluation

Question 16

A company uses a foundation model on Amazon Bedrock for generating responses to customer inquiries. Occasionally, the model returns incomplete responses or errors. The application needs to handle these failures gracefully and provide a fallback. Which strategy is most effective?

Accepted Answer

Catch errors and fall back to a simpler model or a pre-defined static response.

Question 17

A healthcare application uses Amazon Bedrock to analyze patient data. The company must ensure that the foundation model does not retain any patient-specific information after processing. Which measure should be taken?

Accepted Answer

Disable model logging and enable data encryption at rest.

Question 18

A developer is designing a prompt to generate product descriptions. The model often outputs irrelevant or fabricated details. What prompt engineering technique can help reduce such hallucinations?

Accepted Answer

Include explicit instructions to only use information provided in the prompt and ask the model to cite sources.

Question 19

A company needs to automatically retrain a fine-tuned model on new data every month using a pipeline. Which AWS service is best suited for orchestrating this machine learning workflow?

Accepted Answer

Amazon SageMaker Pipelines

Question 20

A developer is building a travel recommendation system that enriches user queries with real-time weather data from an external API before passing them to a foundation model. Which architecture pattern should the developer use?

Accepted Answer

Use Amazon Bedrock Agents with a custom action group that calls the weather API.

Amazon AWS Certified Generative AI Developer - Professional AIP-C01 Free Practice Test - 20 Questions

The free lane is only the front row