Free Practice Questions for Microsoft AI-102 Exam

Scenario Recap:

You have an Azure OpenAI resource with multiple deployed models.

You are using the Chat playground to build a chatbot.

Requirements:

Generate text in concise, formal business language.

Reduce cost of running the model.

Maintain the size of the chatbot history window (i.e., context length).

Step 1: Ensuring concise, formal business language

In the Chat playground setup panel, you can configure System message instructions.

Example:

You are an AI assistant that responds in concise formal business language.

This controls tone, style, and response behavior.

So one setting is: System message / Instructions.

Step 2: Reducing cost while maintaining chat history window

Costs in Azure OpenAI are tied to:

The model used (larger models = more expensive).

The context window size (token usage).

Since we must maintain the size of the chat history window, we cannot shrink context length.

Therefore, to reduce cost: select a smaller/cheaper deployed model (e.g., GPT-3.5 instead of GPT-4).

So the second setting is: Deployment (model selection).

✅ Final Answer:

Configure Deployment → choose a smaller/cheaper model.

Configure System message instructions → specify “concise formal business language.”

Microsoft References:

Azure OpenAI Chat playground

Best practices for reducing token cost

Code Recap (Custom Skill JSON Schema)

{

" @odata.type " : " #Microsoft.Skills.Custom.WebApiSkill " ,

" description " : " My custom skill description " ,

" uri " : " https://contoso-webskill.azurewebsites.net/api/process " ,

" context " : " /document/organizations/* " ,

" inputs " : [

{

" name " : " companyName " ,

" source " : " /document/organizations/* "

}

],

" outputs " : [

{

" name " : " companyDescription "

}

]

}

Statement Analysis

CompanyDescription is available for indexing.

Yes.

The skill produces an output field called companyDescription . Outputs of a custom skill become part of the enrichment tree, which can then be projected into the search index or knowledge store.

The definition calls a web API as part of the enrichment process.

Yes.

This is a Web API skill ( @odata.type: #Microsoft.Skills.Custom.WebApiSkill ) with a uri pointing to https://contoso-webskill.azurewebsites.net/api/process . This means an external web API is called to enrich data.

The enrichment step is called only for the first organization under /document/organizations .

No.

The context is set to " /document/organizations/* " , where the * means the skill runs for each element in the collection. It is not limited to the first element.

Final Answer

CompanyDescription is available for indexing → Yes

The definition calls a web API as part of the enrichment process → Yes

The enrichment step is called only for the first organization under /document/organizations → No

Microsoft References

Custom skills in Azure Cognitive Search

Enrichment tree and skill context

The requirement is to allow User1 to add custom data sources to an Azure OpenAI resource (AI1).

Following the principle of least privilege means granting only the specific rights required.

The Cognitive Services OpenAI Contributor role allows management of Azure OpenAI resources, including attaching custom data sources for RAG (retrieval-augmented generation).

The Cognitive Services Contributor role is broader and grants access to all cognitive services, which violates least privilege.

Search Service Contributor and Search Index Data Contributor are roles for Azure AI Search, not for managing data sources on Azure OpenAI.

Microsoft References:

Azure built-in roles: Cognitive Services OpenAI Contributor

You’re indexing content with Azure AI Search and you want to recognize and retrieve properties from invoices (such as invoice number, vendor, dates, totals, line items). The Azure service designed for structured information extraction from forms and documents is Azure AI Document Intelligence (formerly Form Recognizer).

In an Azure AI Search enrichment pipeline, you add a skillset. To pull structured fields from invoices, include a skill that invokes Document Intelligence’s prebuilt Invoice model. This model already extracts common invoice fields, so you don’t have to train anything—meeting the goal of minimal effort. You can wire this in using the built-in Document Intelligence skill for Search, or as a custom Web API skill that calls the Document Intelligence REST API, and then map the extracted fields into your index for retrieval and filtering.

Why the others are not suitable:

Azure OpenAI is for generative/LLM tasks, not schema-level invoice field extraction.

Azure AI Immersive Reader improves reading accessibility; it doesn’t extract business fields.

Azure Custom Vision is for image classification/object detection, not form/key-value extraction.

Microsoft References (Azure AI Solution)

Azure AI Search skillsets – add AI enrichment and custom skills (Web API skill).

Document Intelligence skill for Azure AI Search (formerly Form Recognizer skill) – integrate extraction into indexing.

Prebuilt Invoice model – fields extracted (vendor, invoice ID/date, totals, taxes, line items, etc.).

Azure AI Document Intelligence overview – key-value extraction and table/line-item parsing from documents.

To generate complete sentences describing an image (for accessibility for visually impaired users), you must call the Describe Image API.

describeImageInStreamAsync analyzes the content of the image and generates human-readable descriptions in natural language sentences.

readInStreamAsync is for OCR (extracting text).

tagImageInStreamAsync is for generating tags (keywords), not full sentences.

analyzeImageByDomainInStreamAsync is for domain-specific models (e.g., celebrities, landmarks).

Correct Answer: B

You are using the Azure OpenAI .NET SDK to send a prompt (“What is Microsoft Azure?”) to a text model deployment. For text generation, you must call the Completions endpoint; hence client.GetCompletions(...) (or the async variant) is correct.

The Response < Completions > object returns one or more choices; the generated text is contained in response.Value.Choices[0].Text , which you can write to the console.

GetEmbeddings would return vector embeddings, not natural-language text, and GetImageGenerations is for DALL·E image generation, so neither meets the requirement to “answer user questions” and print text.

Microsoft References

Azure OpenAI Service (.NET) – Completions quickstart and sample showing GetCompletions and reading Choices[0].Text .

Azure OpenAI Service – Embeddings overview (returns vectors, not text).

Azure OpenAI Service – Image generation overview (DALL·E, not for text answers).

Step 1: Create a new project and load sample documents

Create a new project. Projects store your configurations and settings.

Step 2: Label the sample documents

When you create or open a project, the main tag editor window opens.

Step 3: Train a custom model.

Finally, train a custom model.

The requirements state:

All AI solution responses must have a confidence score ≥ 70%.

If the response confidence score is < 70%, the response must be improved by human input.

Members of the Consultant-Bookkeeper group must be able to process financial documents, which includes performing manual review when the AI confidence is below threshold.

Looking at the provided JSON:

" fields " : {

" ReceiptType " : {

" type " : " string " ,

" valueString " : " Itemized " ,

" confidence " : 0.672

},

" MerchantName " : {

" type " : " string " ,

" valueString " : " Tailwind " ,

" confidence " : 0.913

}

}

ReceiptType.confidence = 0.672 → this is below 0.7.

MerchantName.confidence = 0.913 → this is above 0.7.

Therefore, the correct condition for triggering manual review is:

documentResults.fields. < field > .confidence < 0.7

Now, evaluating the options:

A. documentResults.docType == " prebuilt:receipt "

Always true for receipts, does not check confidence. Not correct.

B. documentResults.fields.*.confidence < 0.7

This is the correct general expression: trigger manual review whenever any field confidence is below 0.7.

C. documentResults.fields.ReceiptType.confidence > 0.7

This would bypass manual review when ReceiptType has high confidence. The requirement is to trigger review when confidence < 0.7, so this is the opposite.

D. documentResults.fields.MerchantName.confidence < 0.7

This only checks one field (MerchantName). In the JSON, MerchantName has confidence 0.913 ( > 0.7), so this condition would not trigger, but ReceiptType clearly needs review. Too narrow.

Correct Answer: B. documentResults.fields.*.confidence < 0.7

Microsoft References

Azure AI Document Intelligence – Confidence scores

Human-in-the-loop for Document Intelligence

Can search content in different formats, including video

Audio and video insights (multi-channels). When indexing by one channel, partial result for those models will be available.

Keywords extraction: Extracts keywords from speech and visual text.

Named entities extraction: Extracts brands, locations, and people from speech and visual text via natural language processing (NLP).

Topic inference: Makes inference of main topics from transcripts. The 2nd-level IPTC taxonomy is included.

Artifacts: Extracts rich set of " next level of details " artifacts for each of the models.

Sentiment analysis: Identifies positive, negative, and neutral sentiments from speech and visual text.

C: Analyze Receipt - Get Analyze Receipt Result.

Query the status and retrieve the result of an Analyze Receipt operation.

Request URL: https://{endpoint}/formrecognizer/v2.0-preview/prebuilt/receipt/analyzeResults/{resultId}

E: POST {Endpoint}/vision/v3.2/read/analyze

Use this interface to get the result of a Read operation, employing the state-of-the-art Optical Character Recognition (OCR) algorithms optimized for text-heavy documents.

Scenario: Contoso plans to develop a document processing workflow to extract information automatically from PDFs and images of financial documents

The document processing solution must be able to process standardized financial documents that have the following characteristics:

- Contain fewer than 20 pages.

- Be formatted as PDF or JPEG files.

- Have a distinct standard for each office.

*The document processing solution must be able to extract tables and text from the financial documents.

The document processing solution must be able to extract information from receipt images.