AI-103 Practice Test 30 — 15 Questions

Question 1

You are using Document Intelligence Studio to build a custom extraction model. You need to verify the model quality before deploying. Which metric is most critical?

Accepted Answer

F1 Score. F1 Score is the harmonic mean of precision and recall, so it captures both false positives and false negatives in a single number. For custom extraction models where missing a field and misreading a field are both costly, F1 gives the most balanced view of overall model quality before deployment.

Answer

Recall only

Answer

Precision only

Answer

Model ID

Question 2

You are optimizing an agentic workflow in Azure AI Foundry. The agent needs to solve complex reasoning problems by breaking them down into distinct sub-tasks, executing them iteratively, and reflecting on intermediate results. Which THREE architectural frameworks or patterns support this behavior? (Choose three.)

Accepted Answer

Reflection pattern where the agent critiques its own intermediate outputs and refines them.. Option A is correct because the Reflection pattern explicitly has the agent generate an initial output, critique it, and iteratively refine intermediate results, which directly supports the required self-evaluation and improvement loop. Option B is correct because ReAct interleaves reasoning traces (thoughts) with actions/tool calls and observations, enabling iterative sub-task execution and dynamic adjustment based on intermediate results. Option E is correct because Plan-and-Solve first decomposes the problem into a complete multi-step plan and then executes the steps, matching the need to break complex reasoning into distinct sub-tasks before iterating through them. Option C is not appropriate because static hardcoded IF-ELSE trees do not use model inference and cannot adaptively reason or reflect on intermediate outputs. Option D is not appropriate because single-turn stateless completions with temperature 0 provide no iterative loop, no memory of prior steps, and no mechanism for reflection or tool execution.

Answer

Static hardcoded IF-ELSE decision trees without model inference.

Answer

Single-turn stateless completion calls with temperature set to zero.

Question 3

When designing an evaluation pipeline in Azure AI Foundry for generative AI applications, which TWO metrics are commonly used to assess safety and risk? (Choose TWO)

Accepted Answer

Sexual content and violence safety metrics. Option C is correct because Azure AI Foundry's safety evaluation metrics specifically include sexual content and violence severity scores, which quantify harmful content categories in generated outputs and are part of the built-in risk and safety evaluators. Option E is correct because hate speech and unfairness detection are also standard safety metrics in Azure AI Foundry, covering hateful or discriminatory language and bias/unfairness assessments of model responses. These metrics align with the Azure AI Content Safety categories (hate, violence, sexual, self-harm) plus fairness evaluations used to assess generative AI risk. Options A and B are incorrect because database transaction log growth rate and storage account geo-replication latency are infrastructure/storage performance measures, not safety or risk metrics for generative AI. Option D is incorrect because network bandwidth throughput in megabits per second is a networking performance metric, unrelated to evaluating content safety or fairness risks.

Answer

Database transaction log growth rate

Answer

Storage account geo-replication latency

Answer

Network bandwidth throughput in megabits per second

Question 4

You are building an agent in Azure AI Foundry that requires access to a private internal database. Which component should you configure to ensure the agent retrieves data securely without exposing the database endpoint publicly?

Accepted Answer

Azure Private Link and Managed Virtual Network. Azure Private Link combined with a Managed Virtual Network ensures that the agent's traffic to the private internal database traverses the Azure backbone via a private endpoint, never exposing the database to the public internet. This is the standard secure pattern for private data access in Azure AI Foundry. It provides network-level isolation and prevents data exfiltration.

Answer

Public IP allowlisting

Answer

Entra ID Application Proxy

Answer

Shared Access Signatures (SAS)

Question 5

You want to add a new tool to an agent. Which step is required to link the tool to the model?

Accepted Answer

Register the tool in the agent configuration within Azure AI Foundry.. In Azure AI Foundry, tools are exposed to an agent through the agent's configuration — you register the tool (function, Azure AI Search, Bing grounding, OpenAPI, etc.) as a tool definition so the orchestrator can pass its schema to the model and invoke it during a run. The model only knows about tools that are declared in the agent/run configuration; there is no implicit discovery from the environment. Registering the tool in the agent configuration is therefore the required linking step.

Answer

Hardcode the tool logic in the system prompt.

Answer

Add the tool code to the global Python environment.

Answer

Update the content safety policy.

Question 6

You are implementing cross-domain document extraction where incoming PDFs contain complex nested tables with merged cells. You need to extract these tables accurately into structured JSON. Which capability of Azure AI Document Intelligence should you rely on?

Accepted Answer

The tables collection in the analyze result, which provides rowSpan, colSpan, and cell bounding regions.. The tables collection in the Document Intelligence analyze result is the correct capability because it returns each detected table with its cells, and each cell includes rowIndex, columnIndex, rowSpan, colSpan, content, and boundingRegions/polygon coordinates. This is exactly what you need to reconstruct merged cells and nested table structure into JSON. The other collections do not preserve the grid geometry required for merged cells.

Answer

The documents collection, which converts tables into Markdown format automatically.

Answer

The keyValuePairs collection, which automatically flattens table rows into key-value pairs.

Answer

The styles collection, which identifies table gridlines through computer vision edge detection.

Question 7

You are designing a disaster recovery and high availability strategy for an enterprise Azure OpenAI deployment. Which THREE architectural components or practices should you include? Each correct answer presents part of the solution.

Accepted Answer

Configure Azure Monitor metric alerts to detect latency spikes or availability drops.. Option A is correct because Azure Monitor metric alerts on latency and availability metrics provide the observability needed to detect degradation of an Azure OpenAI resource and trigger operational or automated responses before users are severely impacted. Option B is correct because Azure OpenAI returns transient errors such as HTTP 429 (rate limiting) and 5xx responses, so client-side retry with exponential backoff and failover to a secondary endpoint/region is essential for resilience and continuity. Option D is correct because deploying secondary Azure OpenAI resources in a paired Azure region enables regional redundancy and supports failover if the primary region becomes unavailable, which is a core HA/DR pattern. Option C is not correct because Azure Backup does not snapshot live model memory states or the internal runtime state of an Azure OpenAI model; it is for supported Azure resource data, not model inference memory. Option E is not correct because Azure OpenAI models run as a managed cloud service and cannot be executed offline on client laptops, and storing model weights locally is neither supported nor part of an Azure HA/DR design.

Answer

Rely exclusively on Azure Backup to snapshot live model memory states.

Answer

Store all model weights locally on client laptops for offline execution.

Question 8

Which THREE types of tools can be integrated into an Azure AI Agent?

Accepted Answer

Azure AI Search indexes.. Azure AI Agents support tool integration through several mechanisms, and Azure AI Search indexes (A) are correct because they can be attached as knowledge/retrieval tools, enabling the agent to ground responses in indexed enterprise data via the Azure AI Search service. Custom Python functions (B) are correct because agents can call user-defined functions (function calling/tool use) to execute custom logic, calculations, or data processing at runtime. External REST APIs (D) are correct because agents can invoke OpenAPI-specified or HTTP-based tools to reach external services and return results into the conversation. Hard drive file systems (C) are not a supported agent tool type, since agents run as cloud services without direct access to a user's local disk, and operating system kernel drivers (E) are not integrable tools because agents operate at the application/service layer, not the OS kernel level.

Answer

Hard drive file systems (local).

Answer

Operating system kernel drivers.

Question 9

You are creating a custom prompt template in Azure AI Foundry. You want to insert the user's input safely while preventing prompt injection vulnerabilities. How should you design the template structure?

Accepted Answer

Use explicit XML tags or markdown delimiters to separate system instructions from user inputs. Wrapping user input in explicit XML tags or markdown delimiters (e.g., ...) creates a clear structural boundary between trusted system instructions and untrusted user content. This lets the model distinguish instruction context from data context, which is the recommended mitigation for prompt injection in Azure AI Foundry prompt templates. It preserves the user's actual input while reducing the chance that injected text is interpreted as a system directive.

Answer

Remove all spaces and punctuation from the user input

Answer

Concatenate user input directly into the middle of the system prompt string

Answer

Encode all user inputs using Base64 before sending to the model

Question 10

You are reviewing the JSON response of a sentiment analysis API call from the Azure AI Language service. Which THREE properties or confidence scores are returned in the standard response payload for a given document?

Accepted Answer

Sentence-level sentiment and confidence scores. Option A is correct because the Azure AI Language sentiment analysis response includes a 'sentences' array where each sentence carries its own 'sentiment' label and 'confidenceScores' (positive, neutral, negative), enabling sentence-level analysis. Option B is correct because every document and sentence in the payload contains a 'confidenceScores' object with numeric values for the positive, neutral, and negative classes. Option D is correct because the top-level document object returns an overall 'sentiment' value that can be positive, negative, neutral, or mixed (mixed applies when sentence sentiments conflict). Option C is incorrect because sentiment analysis does not perform translation; that is a separate Azure AI Translator capability. Option E is incorrect because grammar and spelling correction is not part of the sentiment analysis response payload.

Answer

Exact translation of the text into English

Answer

Grammar and spelling correction suggestions

Question 11

You are auditing an Azure AI Document Intelligence solution for data privacy compliance. Documents contain sensitive Personally Identifiable Information (PII). When using prebuilt models or custom models, how is data handled by Microsoft regarding model training?

Accepted Answer

Customer data and custom model training data are encrypted at rest and in transit, and are never used to train Microsoft foundation models.. Microsoft's Azure AI Document Intelligence service treats customer data with strict privacy controls: all uploaded documents and custom model training data are encrypted at rest (AES-256) and in transit (TLS 1.2+), and Microsoft contractually commits that customer data is never used to train or improve Microsoft's foundation models. This aligns with Microsoft's Responsible AI and data privacy commitments, ensuring that sensitive PII remains isolated to the customer's resource.

Answer

Training data uploaded for custom models is made publicly available in the Azure AI gallery after 30 days.

Answer

Data privacy controls require you to deploy a local container, as cloud APIs store training documents indefinitely.

Answer

All submitted documents are automatically used to retrain the global prebuilt models overnight.

Question 12

You are designing a RAG solution in Azure AI Foundry where sensitive enterprise documents must be retrieved. Users should only see search results corresponding to their security clearance level. How should you implement document-level access control in Azure AI Search?

Accepted Answer

Store user security groups in document metadata and apply OData filter expressions during search queries. Document-level access control in Azure AI Search is implemented by storing security identifiers (user or group IDs) in document metadata fields and applying OData filter expressions at query time to restrict results to documents the caller is authorized to see. The application passes the user's group memberships as filter parameters, and Azure AI Search returns only matching documents. This enforces security trimming at the retrieval layer, which is essential for RAG scenarios where the LLM must not receive unauthorized content.

Answer

Configure Microsoft Entra ID authentication on the Azure OpenAI endpoint

Answer

Encrypt the search index using customer-managed keys (CMK) in Azure Key Vault

Answer

Use Azure AI Content Safety to redact sensitive keywords before indexing

Question 13

Your enterprise customer service department receives thousands of support emails daily. Management wants a system that ingests long support tickets and automatically outputs a concise summary highlighting the most critical issues. Which Azure AI Language service feature should you implement?

Accepted Answer

Text summarization (Abstractive or Extractive). Text summarization in Azure AI Language is designed to condense long documents into concise summaries, exactly matching the requirement to ingest long support tickets and output a concise summary highlighting critical issues. Both abstractive (generates new sentences) and extractive (selects key sentences) modes are available, and the service can handle large volumes of text. The other features focus on extracting specific elements (entities, key phrases, opinions) rather than producing a coherent summary.

Answer

Named Entity Recognition (NER)

Answer

Key phrase extraction

Answer

Opinion mining

Question 14

You are building an agent in Azure AI Foundry using a prompt flow. You need to ensure that the agent only uses company-approved documentation for its responses. Which feature should you implement to meet this requirement?

Accepted Answer

Configure Grounding with data. Grounding with data (e.g., Azure AI Search index, Azure Blob, or other approved data sources) constrains the model's responses to a curated knowledge base, ensuring the agent only answers from company-approved documentation. This is the standard RAG pattern in Azure AI Foundry and directly satisfies the 'only approved documentation' requirement.

Answer

Enable prompt tuning

Answer

Implement system message constraints

Answer

Deploy a fine-tuned model

Question 15

You are using Azure AI Vision to analyze an image. You want to receive the output in Spanish instead of English. Which query parameter should you add to your request?

Accepted Answer

language=es. Azure AI Vision's Analyze Image API accepts a `language` query parameter that specifies the language for the returned results, such as captions and tags. Setting `language=es` returns the output in Spanish.

Answer

translation=spanish

Answer

output-lang=es

Answer

locale=es-ES