Free Microsoft AI-300 Practice Exam with Questions & Answers | Set: 2

Microsoft ' s RAG optimization guidance identifies two high-impact, low-cost improvements for retrieval quality that do not require changing the embedding model. First, tuning chunk size and overlap (option A): chunk size determines how much context each retrieved piece contains — too large and irrelevant content dilutes the signal; too small and answers may be split across chunks. Adjusting these parameters requires only re-indexing the documents with zero additional compute cost. Second, implementing a re-ranker (option B): a re-ranker is a cross-encoder model that takes the top-N retrieved chunks and re-scores them based on their specific relevance to the query, significantly improving precision by filtering out contextually irrelevant chunks. Re-rankers add modest compute cost but are far cheaper than changing the embedding model, which would require re-embedding the entire knowledge base. Increasing token limits (option C) and optimizing embedding vector length (option D) do not address retrieval accuracy without an embedding model change.

Microsoft Learn Reference Topic: Optimize RAG pipelines – Chunk size tuning and re-ranking in Azure AI Search and Azure Machine Learning

LLM model: GPT-4o

Deployment type: Global batch

GPT-4o is the correct model because it is a multimodal model capable of processing both text and image inputs while generating text-based responses. This directly satisfies the requirement to generate content based on a combination of text and images. DALL-E is primarily designed for image generation rather than general multimodal language processing, while Whisper is designed for speech recognition and audio transcription.

Global batch is the correct deployment type because Microsoft specifically positions it for large-scale, high-volume asynchronous processing . Global Batch accepts large groups of requests and processes them using a separate enqueued-token quota. Critically, Microsoft states that this separate quota helps ensure that batch processing does not disrupt online workloads , which directly matches the scenario. It also provides a target turnaround of approximately 24 hours and is optimized for use cases such as large-scale content generation.

Standard deployments are better suited to interactive or lower-volume inference. Global provisioned deployments provide reserved predictable throughput, but they do not provide the specific workload isolation benefit described for asynchronous Global Batch processing.

Study Guide Reference: Design and implement a GenAIOps infrastructure — multimodal model selection, Azure OpenAI deployment types, batch inference, throughput management, and workload isolation.

To ensure runtime dependencies are consistent and reusable, first create a conda.yaml or requirements.txt file that lists all Python packages and system libraries required by your training code — this file is the single source of truth for your runtime. Next, create an Environment object using the Azure ML Python SDK v2 with a name and reference to the conda.yaml file, specifying the base Docker image. Then register the Environment by calling ml_client.environments.create_or_update, which publishes it to the workspace registry with an auto-incremented version. Finally, reference the registered environment by name and version in all pipeline job steps. Azure ML will build or retrieve the cached Docker image and use it as the execution container. This approach means updating dependencies only requires modifying the conda.yaml and registering a new version — training code remains unchanged.

Microsoft Learn Reference Topic: Create and manage Azure Machine Learning environments – Reusable curated environments

Correct sequence:

Create a data asset.

Create a Foundry Search resource.

Create a vector index.

Create a flow.

First, create a data asset so that the locally stored text files are available within the project as a managed data source. The content must then be made searchable for Retrieval-Augmented Generation.

Next, create a Foundry Search resource to provide the search infrastructure that will store and serve the indexed document content. Microsoft documents Azure AI Search as a supported vector store for Foundry and Azure Machine Learning RAG workloads.

Third, create a vector index from the source documents. During vector-index creation, the content is processed into chunks and embeddings, which enables semantic similarity retrieval. Microsoft documents that vector indexes can be created from local files, folders, or registered data assets and then consumed by an Index Lookup operation.

Finally, create a flow and configure it to query the vector index and supply the retrieved context to the connected Azure OpenAI model. Microsoft specifically documents adding an existing vector index to a prompt flow through the Index Lookup tool.

Create a connection is the unused action in this question because the scenario already specifies a connected Azure OpenAI Service model.

Study Guide Reference: Design and implement a GenAIOps infrastructure — RAG architecture, project data, vector indexing, Azure AI Search, prompt flow, and grounding with enterprise content.

Microsoft ' s documentation on Azure Machine Learning model monitoring describes a layered alerting architecture. At the base layer, Azure ML model monitors compute drift, prediction, and data quality metrics on a scheduled basis and publish results as Azure Monitor metrics. To notify stakeholders, you create an Azure Monitor alert rule that watches these metrics and fires an action group when a threshold is breached. Action groups support email, SMS, push notifications, and webhook calls. To initiate automated retraining, the webhook call in the action group targets an Azure ML pipeline ' s REST endpoint, triggering a retraining run. Alternatively, Azure Event Grid subscriptions on AML workspace events can route model-quality events to Azure Functions that start pipelines. The separation of monitoring, alerting, notification, and remediation is intentional, allowing each component to be updated independently.

Microsoft Learn Reference Topic: Set up model monitoring for data and model quality – Azure Machine Learning model monitoring

B and C provide the strongest architecture for reusable and predictable prompts. Separating persistent application instructions from dynamic user input establishes a clear instruction hierarchy . The stable portion of the prompt can define application behavior while channel-specific or request-specific content remains variable. This reduces coupling between user wording and the behavioral rules of the assistant and makes the prompt easier to reuse, version, test, and modify.

The system prompt should establish the model ' s role, behavioral boundaries, tone, communication style, response format, and other persistent rules. Microsoft specifically describes system messages as high-priority instructions used to define an assistant ' s role and boundaries and to set tone and communication style.

Fine-tuning for every conversational variation is unnecessarily expensive and contradicts the requirement to remain adaptable without retraining. Increasing temperature increases randomness and therefore works against consistent responses. Although repeating important instructions can sometimes mitigate recency effects, Microsoft treats this as an optional prompt-engineering technique to experiment with, not the primary architectural technique for producing reusable prompts.

Study Guide Reference: Design and implement a GenAIOps infrastructure — design and develop prompts, prompt management, prompt variants, system instructions, and reusable prompt structures.

===============

Imagine a hospital analytics firm with data scientists who kick off dozens of training jobs throughout the week. During peak hours, five jobs compete for the same GPU cluster and fail or queue for hours. On quiet nights, that cluster sits completely idle, burning money. Managed compute targets with autoscaling solve both problems: the cluster scales out automatically when multiple jobs arrive simultaneously and scales back to zero when idle. Option A (single shared cluster) is exactly the resource-contention problem Fabrikam already has. Option B (fixed-size cluster) wastes money during off-peak hours. Option C (dedicated per-experiment clusters) eliminates contention but is prohibitively expensive for a cost-conscious healthcare firm. Autoscaling managed compute is the cloud-native answer to variable workload demand.

Microsoft Learn Reference Topic: Azure Machine Learning compute targets – Configure autoscale for compute clusters

Domain specialization: Supervised fine-tuning

Poor response accuracy of a RAG-based solution: Apply prompt engineering

For domain specialization , Supervised Fine-Tuning (SFT) is the correct first strategy. Microsoft describes SFT as the foundational fine-tuning technique for training a model from labeled input-output pairs , and specifically identifies domain specialization as one of its principal use cases. Microsoft also recommends starting with SFT for most customization projects because it supports task specialization, instruction following, style, and domain-specific behavior. Fabrikam already possesses evaluation data containing input-output pairs, which aligns directly with the SFT data model.

For poor RAG response accuracy , the first action is prompt engineering . The case explicitly requires advanced fine-tuning only when prompt engineering is insufficient. In a RAG system, prompt engineering determines how the model interprets retrieved context, constrains answers to grounding information, handles missing evidence, and formats responses. Microsoft notes that inadequate RAG prompting can produce false or incomplete answers even when retrieval returns appropriate content.

DPO is primarily appropriate for alignment using preferred versus non-preferred responses, while RFT targets complex reward-based reasoning optimization. Neither is the initial technique for domain specialization in this requirement.

Study Guide Reference: Optimize generative AI systems and model performance — prompt engineering, RAG optimization, supervised fine-tuning, preference optimization, and model customization strategy.

Managed online endpoints with traffic splitting are the best fit because Azure Machine Learning supports multiple deployments behind a single online endpoint and allows production traffic to be distributed between them. This enables a blue-green or progressive rollout strategy : deploy the new model version alongside the current version, validate it independently, send a small percentage of production traffic to it, monitor behavior, and then gradually increase traffic. If problems occur, traffic can be redirected immediately to the previous deployment, providing a practical rollback mechanism.

This approach also aligns with Fabrikam’s requirement to favor managed services and automation because the team has limited Azure DevOps experience. Managed online endpoints handle much of the serving, scaling, security, and monitoring infrastructure automatically and support both traffic splitting and mirrored traffic for safer validation.

Option B can also implement blue-green deployment, but AKS introduces substantially more infrastructure and operational management than Fabrikam requires. Option A would require custom deployment, routing, scaling, and rollback logic. Option D is intended for asynchronous batch inference rather than staged real-time production serving.

Study Guide Reference: Implement machine learning model lifecycle and operations — managed online endpoints, multiple deployments, traffic splitting, blue-green deployment, staged rollout, monitoring, and rollback.