AI-102: Designing and Implementing a Microsoft Azure AI Solution

a. Select the appropriate Azure AI Foundry services

• Select the appropriate service for a generative AI solution
• Select the appropriate service for a computer vision solution
• Select the appropriate service for a natural language processing solution
• Select the appropriate service for a speech solution
• Select the appropriate service for an information extraction solution
• Select the appropriate service for a knowledge mining solution

b. Plan, create and deploy an Azure AI Foundry service

• Plan for a solution that meets Responsible AI principles
• Create an Azure AI resource
• Choose the appropriate AI models for your solution
• Deploy AI models using the appropriate deployment options
• Install and utilize the appropriate SDKs and APIs
• Determine a default endpoint for a service
• Integrate Azure AI Foundry Services into a continuous integration and continuous delivery (CI/CD) pipeline
• Plan and implement a container deployment

c. Manage, monitor, and secure an Azure AI Foundry Service

• Monitor an Azure AI resource
• Manage costs for Azure AI Foundry Services
• Manage and protect account keys
• Manage authentication for an Azure AI Foundry Service resource

d. Implement AI solutions responsibly

• Implement content moderation solutions
• Configure responsible AI insights, including content safety
• Implement responsible AI, including content filters and blocklists
• Prevent harmful behavior, including prompt shields and harm detection
• Design a responsible AI governance framework

a. Build generative AI solutions with Azure AI Foundry

• Plan and prepare for a generative AI solution
• Deploy a hub, project, and necessary resources with Azure AI Foundry
• Deploy the appropriate generative AI model for your use case
• Implement a prompt flow solution
• Implement a RAG pattern by grounding a model in your data
• Evaluate models and flows
• Integrate your project into an application with Azure AI Foundry SDK
• Utilize prompt templates in your generative AI solution

b. Use Azure OpenAI in Foundry Models to generate content

• Provision an Azure OpenAI in Foundry Models resource
• Select and deploy an Azure OpenAI model
• Submit prompts to generate code and natural language responses
• Use the DALL-E model to generate images
• Integrate Azure OpenAI into your own application
• Use large multimodal models in Azure OpenAI
• Implement an Azure OpenAI Assistant

c. Optimize and operationalize a generative AI solution

• Configure parameters to control generative behavior
• Configure model monitoring and diagnostic settings, including performance and resource consumption
• Optimize and manage resources for deployment, including scalability and foundational model updates
• Enable tracing and collect feedback
• Implement model reflection
• Deploy containers for use on local and edge devices
• Implement orchestration of multiple generative AI models
• Apply prompt engineering techniques to improve responses
• Fine-tune an generative model

a. Create custom agents

• Understand the role and use cases of an agent
• Configure the necessary resources to build an agent
• Create an agent with the Azure AI Foundry Agent Service
• Implement complex agents with Semantic Kernel and Autogen
• Implement complex workflows including orchestration for a multi-agent solution, multiple users, and autonomous capabilities
• Test, optimize and deploy an agent

a. Analyze images

• Select visual features to meet image processing requirements
• Detect objects in images and generate image tags
• Include image analysis features in an image processing request
• Interpret image processing responses
• Extract text from images using Azure AI Vision
• Convert handwritten text using Azure AI Vision

b. Implement custom vision models

• Choose between image classification and object detection models
• Label images
• Train a custom image model, including image classification and object detection
• Evaluate custom vision model metrics
• Publish a custom vision model
• Consume a custom vision model
• Build a custom vision model code first

c. Analyze videos

• Use Azure AI Video Indexer to extract insights from a video or live stream
• Use Azure AI Vision Spatial Analysis to detect presence and movement of people in video
• 
  

a. Analyze and translate text

• Extract key phrases and entities
• Determine sentiment of text
• Detect the language used in text
• Detect personally identifiable information (PII) in text
• Translate text and documents by using the Azure AI Translator service

b. Process and translate speech

• Integrate generative AI speaking capabilities in an application
• Implement text-to-speech and speech-to-text using Azure AI Speech
• Improve text-to-speech by using Speech Synthesis Markup Language (SSML)
• Implement custom speech solutions with Azure AI Speech
• Implement intent and keyword recognition with Azure AI Speech
• Translate speech-to-speech and speech-to-text by using the Azure AI Speech service

c. Implement custom language models

• Create intents, entities, and add utterances
• Train, evaluate, deploy, and test a language understanding model
• Optimize, backup, and recover language understanding model
• Consume a language model from a client application
• Create a custom question answering project
• Add question-and-answer pairs and import sources for question answering
• Train, test, and publish a knowledge base
• Create a multi-turn conversation
• Add alternate phrasing and chit-chat to a knowledge base
• Export a knowledge base
• Create a multi-language question answering solution
• Implement custom translation, including training, improving, and publishing a custom model

a. Implement an Azure AI Search solution

• Provision an Azure AI Search resource, create an index, and define a skillset
• Create data sources and indexers
• Implement custom skills and include them in a skillset
• Create and run an indexer
• Query an index, including syntax, sorting, filtering, and wildcards
• Manage Knowledge Store projections, including file, object, and table projections
• Implement semantic and vector store solutions

b. Implement an Azure AI Document Intelligence solution

• Provision a Document Intelligence resource
• Use prebuilt models to extract data from documents
• Implement a custom document intelligence model
• Train, test, and publish a custom document intelligence model
• Create a composed document intelligence model

c. Extract information with Azure AI Content Understanding

• Create an OCR pipeline to extract text from images and documents
• Summarize, classify, and detect attributes of documents
• Extract entities, tables, and images from documents
• Process and ingest documents, images, videos, and audio with Azure AI Content Understanding