AI-102 Practice Test 49 — 15 Questions

Question 1

You are deploying a generative AI chat application that calls an Azure OpenAI model deployment. The application must stream partial completions to the browser as tokens are produced and must reduce perceived latency for long answers. Which two request options should you configure? (Choose two.)

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Set stream to true on the chat completions request. Streaming has two cooperating parts: the request must set stream to true so the service emits incremental deltas, and the client must read those server-sent events and forward them to the browser. Together they let users see text as it is generated. Parameters that alter sampling, penalties, or probabilities change content or add metadata but never change delivery timing.

Answer

Increase the frequency_penalty value to discourage repeated tokens

Answer

Set the logprobs parameter to return token-level probabilities

Answer

Set n to a value greater than one to generate multiple candidate completions

Question 2

You plan to deploy a custom named entity recognition (NER) project in Azure AI Language. The project must extract supplier names and contract identifiers from procurement documents. You need to prepare the project so that the model can be trained and evaluated before deployment. Which two actions should you perform? (Choose two.)

Accepted Answer

Upload and label training documents by selecting the text spans for SupplierName and ContractId entities.. To train a custom NER model, you create a project of the custom named entity recognition type with the correct language, then upload and label documents by marking the spans for each entity such as SupplierName and ContractId. Training and evaluation follow, and only then can the model be deployed for extraction.

Answer

Define a list of intents that correspond to each procurement document type.

Answer

Create a question answering knowledge base that contains the procurement documents.

Answer

Enable opinion mining so the model can detect sentiment around supplier mentions.

Question 3

You are creating an agent in Microsoft Foundry that answers questions about internal HR policies. The policy documents are already indexed in Azure AI Search, and you want the fastest path to a working agent without writing retrieval code. What should you do first?

Accepted Answer

Create the agent, then connect the existing Azure AI Search index as a knowledge source for the agent.. Connecting the existing Azure AI Search index as a knowledge source gives the agent retrieval over the HR policies with no custom retrieval code. It reuses the index the organization already maintains, so content updates flow through automatically and the agent is functional with minimal configuration.

Answer

Upload the policy documents to the agent's file storage and enable file search.

Answer

Export the policies to CSV and attach them as a code interpreter file.

Answer

Build an Azure Functions app that queries the index and expose it as a function tool.

Question 4

You are deploying a generative AI feature that drafts marketing copy with an Azure OpenAI GPT-4o deployment. The application must stream partial responses to the browser so users see text as it is generated, rather than waiting for the full completion. What should you enable in your API call?

Accepted Answer

Set the stream parameter to true and consume the server-sent event chunks returned by the chat completions endpoint.. Streaming in Azure OpenAI is opt-in per request: setting the stream flag makes the chat completions endpoint return incremental deltas over server-sent events, which the browser can render progressively. Polling, batch processing, and quota changes all describe non-interactive or throughput-oriented behaviors that do not deliver partial tokens as they are generated.

Answer

Enable the asynchronous batch API and retrieve results from the output file once the job finishes.

Answer

Set a high max_tokens value and poll the operation status endpoint until the completion is marked complete.

Answer

Increase the deployment's tokens-per-minute quota so responses are generated faster.

Question 5

A developer is building a generative AI application with Azure OpenAI Service. The application must stream partial responses to users as tokens are generated, rather than waiting for the complete response. Which parameter should the developer set in the chat completion request?

Accepted Answer

Set 'stream' to true.. Streaming is controlled by the stream parameter, which when true causes the service to emit incremental delta chunks over a persistent connection. This lets the application display tokens as they are generated, reducing perceived latency while keeping the same model and prompt configuration.

Answer

Set 'temperature' to 0.

Answer

Set 'max_tokens' to a large value.

Answer

Set 'n' to a value greater than 1.

Question 6

You are implementing a conversational language understanding (CLU) project in Azure AI Language. Your utterances include entities that are sometimes a single word and sometimes a multi-word phrase, such as 'New York' and 'San Francisco'. You need the model to correctly capture these multi-word entities during training and prediction. Which entity component type should you use?

Accepted Answer

Learned entity component. Learned entity components in CLU are trained from labeled utterances, allowing the model to identify entity spans based on surrounding context. This is the correct choice for multi-word entities such as city names that vary in phrasing and cannot be captured reliably by exact-match lists or fixed regex patterns.

Answer

Prebuilt entity component

Answer

Regex entity component

Answer

List entity component

Question 7

A media company is building an Azure AI Foundry agent that generates summaries of news articles. The agent uses a tool to fetch articles from an internal CMS. The company wants to ensure the agent respects content usage rights and does not summarize articles that are marked as restricted. The agent must also log every article it accesses for auditing. Which two actions should the team take? (Choose two.)

Accepted Answer

Enable diagnostic logging on the Azure AI Foundry resource to capture all tool calls and their responses.. To respect usage rights, the CMS tool should filter articles based on the usage rights field before they reach the agent, ensuring only permissible content is summarized. To audit access, diagnostic logging on the Azure AI Foundry resource captures all tool calls and responses. Together, these actions enforce policy at the data layer and provide a verifiable audit trail without exposing restricted content to the model.

Answer

Instruct the agent in its system message to ignore articles that are marked restricted.

Answer

Use a content filter to block articles that contain certain keywords associated with restricted content.

Answer

Store a copy of every article in the agent's conversation history for later review.

Question 8

You are building a generative AI application with Azure OpenAI Service that summarizes long legal contracts. Legal reviewers report that the summaries omit obligations buried in the middle of very long documents, even though the documents fit within the model's maximum context window. You need to improve recall of those mid-document clauses while minimizing added latency. What should you do first?

Accepted Answer

Split the contract into overlapping chunks, summarize each chunk, then summarize the combined chunk summaries.. Long inputs suffer from uneven attention, so obligations in the middle can be missed even when they fit the context window. Chunking with overlap and summarizing hierarchically places every clause in a short, high-attention context and preserves clauses that cross boundaries. Raising temperature, enlarging the window, or adding instructions does not change how attention is distributed, so those approaches leave the recall gap unresolved.

Answer

Add a system message instructing the model to read the entire document carefully before summarizing.

Answer

Increase the model's temperature setting so the model explores more of the document content.

Answer

Switch the deployment to a model with a larger maximum context window and resend the full contract.

Question 9

You are planning an Azure AI solution that will use several Azure AI services, including Azure AI Vision and Azure AI Language. The solution must be deployed to multiple regions and must provide a single endpoint and key for all the services. Which Azure resource type should you create?

Accepted Answer

An Azure AI services multi-service account. A multi-service account is designed to expose several Azure AI services through one endpoint and one key. It supports deployment in multiple regions while keeping a consistent management and authentication model. Separate resources or an API Management layer would still leave you managing per-service credentials, so they do not satisfy the single-key requirement.

Answer

An Azure API Management instance with a backend for each service

Answer

An Azure Machine Learning workspace with online endpoints

Answer

A separate Azure AI services resource for each service in each region

Question 10

You are developing a solution that uses Azure AI Language's custom named entity recognition (NER) to extract product names from technical support tickets. After training a model, you notice it performs well on the training set but poorly on new tickets. You need to improve the model's ability to generalize. What should you do?

Accepted Answer

Add more labeled examples that include variations in phrasing, context, and entity placement.. Poor generalization on new data indicates overfitting. The best remedy is to expand the training set with varied examples that reflect real-world diversity, helping the model learn robust patterns. Increasing epochs or reducing data would exacerbate overfitting, and prebuilt models cannot extract custom entities like product names.

Answer

Reduce the training dataset size to prevent the model from memorizing too much information.

Answer

Increase the number of training epochs to allow the model to learn more from the existing data.

Answer

Switch to a prebuilt entity extraction model, because custom models cannot generalize.

Question 11

You are deploying an Azure AI solution that uses Azure AI Language and Azure AI Vision. You need to ensure that the solution can be monitored for performance and health. You want to collect metrics and logs from these services and analyze them in a central location. What should you use?

Accepted Answer

Azure Monitor. Azure Monitor is the correct choice because it is the comprehensive monitoring solution for Azure resources. It can collect metrics and logs from Azure AI Language and Azure AI Vision, store them in Log Analytics, and enable analysis, alerting, and visualization. The other services provide health advisories or recommendations but do not offer centralized telemetry collection and analysis.

Answer

Azure Resource Health

Answer

Azure Service Health

Answer

Azure Advisor

Question 12

You are designing an Azure AI Search solution that uses an AI enrichment pipeline to extract text and key phrases from scanned documents. You need to ensure the pipeline can process image-heavy PDFs and produce searchable text and key phrases. Which two actions should you include in the skillset? (Choose two.)

Accepted Answer

Add the built-in Key Phrase Extraction skill to identify important phrases from the OCR text.. To process image-heavy PDFs, the OCR skill extracts text from embedded images. Once text is available, the Key Phrase Extraction skill identifies important phrases. Together, these skills enable the pipeline to produce searchable text and key phrases. Other skills like Language Detection, Text Merge, or Entity Recognition serve different purposes and do not fulfill both requirements.

Answer

Add the built-in Language Detection skill to determine the language of each document.

Answer

Add the built-in Text Merge skill to combine multiple text inputs into a single field.

Answer

Add the built-in Entity Recognition skill to extract people, organizations, and locations.

Question 13

You have an Azure AI Search indexer that enriches documents with a custom skill that calls an external API. The custom skill returns a JSON object containing a list of product codes. You need to store these product codes in a collection field named 'productCodes' in the index, and you want the field to be searchable and filterable. What should you do?

Accepted Answer

In the index definition, set the 'productCodes' field to type Collection(Edm.String) and mark it as searchable and filterable. Then in the indexer, map the custom skill output to this field using an output field mapping.. To store multiple product codes from a custom skill, the index field must be a collection of strings. Collection(Edm.String) supports multiple values and can be marked searchable and filterable to meet query requirements. The custom skill's output is part of the enriched document, so an output field mapping is the correct mechanism to project it into the index field. Using a single string field or omitting searchable would not satisfy the scenario.

Answer

In the index definition, set the 'productCodes' field to type Edm.String and mark it as searchable and filterable. Then in the indexer, map the custom skill output to this field using an output field mapping.

Answer

In the index definition, set the 'productCodes' field to type Collection(Edm.String) and mark it as retrievable only. Then in the indexer, map the custom skill output using a field mapping.

Answer

In the index definition, set the 'productCodes' field to type Collection(Edm.String) and mark it as filterable and facetable. Then in the indexer, use a field mapping to map the custom skill output.

Question 14

A company uses Azure Document Intelligence to analyze prebuilt invoices. They need to extract the invoice total and the due date from each invoice. They call the Analyze Invoice operation and receive a result. Which part of the JSON response contains the extracted fields?

Accepted Answer

The documents array, where each document has a fields object with named fields such as InvoiceTotal and DueDate.. The prebuilt invoice model returns a structured JSON with a documents array. Each document includes a fields object that contains named fields such as InvoiceTotal and DueDate, along with their values and confidence scores. This is the correct location to extract specific invoice data.

Answer

The tables array, where each table has cells that contain the extracted values.

Answer

The pages array, where each page has a lines array containing text and bounding boxes.

Answer

The analyzeResult object, which contains a keyValuePairs array with keys and values.

Question 15

You are building a generative AI assistant on Azure OpenAI that must call internal REST APIs to look up order status. You decide to use function calling. Which TWO actions are required to make the assistant reliably invoke the correct API and return a coherent answer? (Choose two.)

Accepted Answer

Execute the function in your application code and send the result back to the model as a tool message so it can generate the final response.. Function calling is a two-part contract. First, the application declares available functions with names, descriptions, and JSON schemas in the tools parameter so the model can choose one and emit structured arguments. Second, the application executes the chosen function and returns the result as a tool message, allowing the model to compose the final answer. Temperature, browsing, and token limits do not enable this loop.

Answer

Set the model temperature to zero so the function call arguments are always deterministic.

Answer

Enable the model's built-in web browsing tool so it can reach the internal APIs directly.

Answer

Increase the max_tokens value to ensure the function call arguments fit in the response.