Azure AI Components and Services
Azure Training Architect II
Artificial Intelligence, or AI, and the Machine Learning that drives it, is one of the most exciting new frontiers of technology being presented in the Cloud today. Microsoft's AI-100 is a certification path and exam that covers the Azure AI services and products available.
The Certification, as described by Microsoft:
Candidates for this exam analyze the requirements for AI solutions, recommend appropriate tools and technologies, and implements solutions that meet scalability and performance requirements.
Candidates translate the vision from solution architects and work with data scientists, data engineers, IoT specialists, and AI developers to build complete end-to-end solutions. Candidates design and implement AI apps and agents that use Microsoft Azure Cognitive Services and Azure Bot Service. Candidates can recommend solutions that use open source technologies.
Candidates understand the components that make up the Azure AI portfolio and the available data storage options.
Candidates implement AI solutions that use Cognitive Services, Azure bots, Azure Search, and data storage in Azure. Candidates understand when a custom API should be developed to meet specific requirements.
The second course covers the various components of an AI or ML based design as well as how the more common interactions work.
Here we outline what can be expected in the second of five courses reviewing the material for the AI-100 Certification.
About the Training Architect
Meet the Training Architect, Dan Sasse: https://www.linkedin.com/in/danielsasse/
Design AI Solutions
Design Solutions that Include One or More Pipelines
Define an AI Application Workflow Process
This lesson centers around defining what an AI application workflow process is and how that changes depending on if an app is based on a custom ML model or a prebuilt offering like a Cognitive Service. Further reading documentation: https://docs.microsoft.com/en-us/azure/machine-learning/service/concept-azure-machine-learning-architecture https://docs.microsoft.com/en-us/azure/machine-learning/service/concept-designer
Design a Strategy for Ingest and Egress Data
In this lesson, we discuss strategies around designing and managing the data ingestion and egression (or outbound flow) from an AI/ML solution. We consider Azure Data Factory as a solution to this challenge. Further reading documentation: https://docs.microsoft.com/en-us/azure/data-factory/introduction https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/data-transformation https://docs.microsoft.com/en-us/azure/azure-databricks/what-is-azure-databricks
Design the Integration Point Between Multiple Workflows and Pipelines
In this lesson, we discuss Azure Logic Apps, Azure Functions, and Azure Function Proxies as solutions to designing integration points between multiple workflows and pipelines. Further reading documentation: https://docs.microsoft.com/en-us/azure/azure-functions/functions-proxies https://docs.microsoft.com/en-us/azure/data-factory/concepts-pipelines-activities
Design Pipelines that use AI Apps
In this lesson, we review what a pipeline is as it relates to application design. We also review how an AI app is presented to a pipeline for inclusion in the overall software development goal. Further reading documentation: https://docs.microsoft.com/en-us/azure/machine-learning/studio/manage-new-webservice https://docs.microsoft.com/en-us/azure/connectors/apis-list
Design Pipelines that Call Azure Machine Learning Models
This lesson dives a little deeper into how the Azure Machine Learning Studio and Machine Learning Service present their predictive models to pipelines that need to call them. Further reading documentation: https://docs.microsoft.com/en-us/azure/machine-learning/service/concept-ml-pipelines https://docs.microsoft.com/en-us/azure/machine-learning/studio/create-endpoint
Select an AI Solution that Meets Cost Constraints
This lesson reviews the factors that come into play when determining the costs of an AI or machine learning solution. We also discuss how to account for the costs in order to accommodate a cost constraint. Further reading documentation: https://azure.microsoft.com/en-us/pricing/details/bot-service/ https://azure.microsoft.com/en-us/pricing/details/cognitive-services/ https://azure.microsoft.com/en-us/pricing/details/machine-learning-studio/
Design Solutions that Uses Cognitive Services
Design Solutions that Use Vision, Speech, Language, Knowledge, Search, and Anomaly Detection APIs
This lesson focuses on the specific, mechanical aspects of the Cognitive Service APIs that an AI-100 Test Taker needs to know for the Exam. Further Reading Documentation: https://docs.microsoft.com/en-us/azure/cognitive-services/kes/overview https://docs.microsoft.com/en-us/azure/#pivot=products&panel=ai
Design Solutions that Implement the Bot Framework
Integrate Bots and AI Solutions
This lesson centers around the Microsoft Bot Framework; it's components, how a bot is tested and communicated with once it is deployed. Further Reading Documentation: https://docs.microsoft.com/en-us/azure/bot-service/bot-service-concept-templates https://docs.microsoft.com/en-us/azure/bot-service/rest-api/bot-framework-rest-connector-api-reference
Design Bot Services that use Language Understanding (LUIS)
In this lesson, we review the Language Understanding Service, discuss the methods with which it can be designed and deployed, and briefly review how it functions in the background and the benefits of using LUIS over another more specific Cognitive Service. Further Reading Documentation: https://docs.microsoft.com/en-us/azure/cognitive-services/luis/what-is-luis https://docs.microsoft.com/en-us/azure/cognitive-services/luis/luis-concept-utterance
Design Bots that Integrate with Channels
This lesson reviews the various channels available to integrate with Bots and how we specify which channel a bot should use. Further Reading Documentation: https://docs.microsoft.com/en-us/azure/bot-service/bot-service-manage-channels https://docs.microsoft.com/en-us/azure/bot-service/bot-service-channels-reference
Integrate Bots with Azure App Services and Azure Application Insights
This lesson touches on the important aspects of Bot Deployments, which are the hosting/compute platform the monitoring, and event logging service. Further Reading Documentation: https://docs.microsoft.com/en-us/azure/azure-monitor/app/app-insights-overview https://docs.microsoft.com/en-us/azure/app-service/
Design the Compute Infrastructure to Support a Solution
Identify Whether to Create a GPU, FPGA, or CPU-Based Solution
This lesson covers the three different compute architectures available for use within Azure. Further Reading Documentation: https://docs.microsoft.com/en-us/azure/machine-learning/service/how-to-deploy-fpga-web-service https://docs.microsoft.com/en-us/azure/machine-learning/service/how-to-deploy-inferencing-gpus https://docs.microsoft.com/en-us/Azure/virtual-machines/windows/sizes
Identify Whether to Use a Cloud-Based, On-Premises, or Hybrid Compute Infrastructure
This lesson highlights the key differences between these three infrastructure options and reviews what is necessary for the AI-100 Exam Taker to keep in mind about them. Further Reading Documentation: https://docs.microsoft.com/en-us/office365/enterprise/hybrid-cloud-overview https://docs.microsoft.com/en-us/azure/application-gateway/overview https://docs.microsoft.com/en-us/azure/service-bus-messaging/service-bus-messaging-overview
Select a Compute Solution that Meets Cost Constraints
This lesson provides an overview of all the different aspects needing to be considered when designing an AI solution where cost management is a primary focus. Further Reading Documentation: https://docs.microsoft.com/en-us/azure/architecture/guide/technology-choices/compute-decision-tree https://docs.microsoft.com/en-us/azure/architecture/guide/technology-choices/compute-overview
Where do you go from here?
Good work! Now: on to the next Course Segment!
Well done on completing Course Segment 2, Components and Services, in our Certification Review of the Azure AI-100. We'd appreciate some feedback before you move on to Segment 3; Workflow and Pipelines.
Take this course and learn a new skill today.
Transform your learning with our all access plan.Start 7-Day Free Trial