AI Bot
AI-200: Azure AI Services Overview Flashcards
Microsoft Azure AI Cloud Developer Associate AI-200 Flashcards
Study our AI-200: Azure AI Services Overview flashcards for the Microsoft Azure AI Cloud Developer Associate AI-200 exam with 30+ flashcards. View as flashcards, a searchable table, or as a fun matching game.
| Front | Back |
| Advantage of using REST APIs | Direct access from any platform and fine grained control over requests and network policies |
| Advantage of using SDKs | Easier integration with error handling retries and authentication helpers |
| Authentication methods for Azure AI services | API keys Azure Active Directory and managed identities for secure access |
| Best practice for production authentication | Use Azure AD and managed identities avoid embedding static keys and enforce least privilege |
| Common pricing tiers for Cognitive Services | Free F0 and Standard S0 for many services with limited quotas on Free |
| Difference between online and batch inference | Online inference serves real time single requests Batch inference handles large scale offline jobs |
| Difference between SDKs and REST APIs | SDKs provide language idiomatic helpers and client objects REST APIs provide raw HTTP endpoints for any language |
| How do quotas affect AI services | Quotas limit requests throughput and resource consumption and can be increased by support request |
| How do you secure REST API calls to Azure AI services | Use HTTPS authorization headers with API keys or OAuth tokens and restrict network access |
| How does pricing vary across regions | Prices differ due to infrastructure costs local taxes and availability which affects cost planning |
| How does region availability affect Azure OpenAI | Azure OpenAI is available in selected regions and may require subscription access or quotas |
| How to request a quota increase | Use the Azure portal support and quotas page to request higher limits for a subscription and region |
| What are Cognitive Services containers | Containerized versions of some Cognitive Services for on prem or edge deployment |
| What are managed identities | Azure managed identities allow resources to authenticate to services without storing credentials |
| What is a pricing tier in Azure AI | Pricing tier determines cost performance feature set and available quota like Free and Standard |
| What is Azure Cognitive Services | A set of pretrained APIs for vision language speech and decision tasks that require minimal ML training |
| What is Azure Machine Learning | A platform for building training deploying and managing custom machine learning models |
| What is Azure OpenAI fine tuning | Customizing model behavior with additional training data or instructions to improve outputs |
| What is Azure OpenAI Service | A managed service providing access to OpenAI models for text and embeddings via Azure |
| What is endpoint key rotation | Process of changing API keys regularly to reduce credential exposure and meet security policies |
| What is endpoint versioning | APIs expose versions to maintain compatibility when introducing new features or breaking changes |
| What is model deployment in Azure ML | Process of packaging a model and exposing it as a prediction endpoint |
| What is the difference between multi service Cognitive Services and single service resources | Multi service unified resource provides many APIs in one Single service resources isolate billing and quotas |
| What is the role of a resource in Azure AI services | An Azure resource groups service configuration billing and quota scope for using an AI service |
| When choose Azure ML over Cognitive Services | When you need full control to build train tune and operationalize custom models |
| When choose Azure OpenAI over Cognitive Services | When you need large language models for complex natural language generation or embeddings |
| When choose Cognitive Services over Azure ML | When you need ready made APIs quickly and do not want to build or train custom models |
| When to use Azure AD over API keys | Use Azure AD for enterprise identity control role based access and easier key rotation |
| When to use containers for Cognitive Services | When you need data residency offline capabilities or low latency at the edge |
| Why region selection matters for AI services | Region affects latency data residency model availability and pricing |
About the Flashcards
Flashcards for the Microsoft Azure AI Cloud Developer Associate exam help you review core Azure AI terminology and service roles, including Azure Cognitive Services, Azure Machine Learning, and Azure OpenAI Service. Cards summarize when to choose each offering, the differences between SDKs and REST APIs, authentication options like API keys, Azure AD and managed identities, and how pricing tiers, quotas, and region selection affect deployments.
They also cover practical operations often tested on the exam: model deployment and endpoints, online versus batch inference, Cognitive Services containers and resource models (multi-service vs single-service), endpoint versioning, fine-tuning Azure OpenAI, and security best practices such as key rotation and securing API calls.
Topics covered in this flashcard deck:
- Azure Cognitive Services
- Azure Machine Learning
- Azure OpenAI Service
- Authentication and identities
- Deployment and inference
- Pricing, quotas, regions