Designing Microsoft Azure AI Solutions

Paths

Designing Microsoft Azure AI Solutions

Authors: James Bannan, Brian Harrison, David Tucker, Ifedayo Bamikole, Jared Rhodes, Sahil Malik

Microsoft Azure offers a robust and powerful AI platform. In order to use it effectively, you will need to understand how the Azure service offering fits together to provide a... Read more

What you will learn:

  • Choosing the appropriate Microsoft Azure services and features for your AI business needs
  • Designing security into the AI solution
  • Applying version control to your AI solution
  • Operationalizing your AI solution as interacting Microsoft Azure services

Pre-requisites

This skill path is intended for learners with experience designing AI solutions, but who are novice with regards to designing them against Microsoft Azure. Prior knowledge of machine learning principles and data science workflows is expected.

Beginner

The courses in this section will give you a rudimentary introduction to the field of Microsoft Azure services that will play a role in your AI solution. In addition, these courses will help you identify the operational considerations of your solution, such as deployment, monitoring, and maintenance.

Choosing the Appropriate Microsoft Azure Services and Features

by James Bannan

Dec 19, 2019 / 2h 0m

2h 0m

Start Course
Description

AI solutions in Microsoft Azure consist of a number of independent resources and services working together holistically to produce a complex solution. As an AI Engineer, In this course, Choosing the Appropriate Microsoft Azure Services and Features, you’ll gain understanding of the processes which are inherent within an AI solution as well as some of the challenges faced by data scientists in the building and training of machine learning models in order to be able to recommend the most appropriate services and features. First, you’ll look at the different types of data that you’re going to need to understand and deal with, including recommending the right services to support different data scenarios. Then, you’ll explore the difference between machine learning models and frameworks, including the different types of machine learning models, incorporating both standard machine learning as well as deep learning. Finally, you’ll explore machine learning pipelines and automated machine learning, including the business and technical challenges that these technologies are designed to overcome. At the end of this course, you will have a basic understanding of the different Azure services.

Table of contents
  1. Course Overview
  2. Understanding Storage for Azure AI Solutions
  3. Understanding Processing in Azure AI Solutions
  4. Understanding Azure AI Services and Models
  5. Understanding Azure AI Components and Endpoints
  6. Understanding Automated Machine Learning

Identifying Security Requirements of an AI Solution

by Brian Harrison

Dec 12, 2019 / 1h 29m

1h 29m

Start Course
Description

Learn what it takes to secure your newly developed and/or deployed AI solution in Microsoft Azure. In this course, Identifying Security Requirements of an AI solution, you will learn foundational knowledge to secure your AI Solution no matter whether you are leveraging Azure Cognitive Services or custom machine learning models. First, you will discover how to leverage Azure Cognitive Services in a secure authenticated way. Next, you will learn what security options are available for any test data that you need to train the models of both Cognitive Services and machine learning models. Finally, you will explore how to secure the deployment of an AI solution using machine learning models. When you are finished with this course, you will have the skills and knowledge of security options within Azure Services needed to secure an AI solution from development to deployment.

Table of contents
  1. Course Overview
  2. Working with Authorization and Authentication
  3. Security of Test Data
  4. Securing an AI Solution Deployment

Intermediate

These courses teach you how to design and operationalize the machine learning components of your AI solution on Microsoft Azure.

Designing Machine Learning Solutions on Microsoft Azure

by David Tucker

Dec 6, 2019 / 1h 41m

1h 41m

Start Course
Description

When working on data science initiatives it can be challenging to gain actionable insights from your data set. In this course, Designing Machine Learning Solutions on Microsoft Azure, you will learn how to leverage Azure's Machine Learning capabilities to greatly increase the chance of success for your data science project. First, you will engage in team workflow and how Microsoft's Team Data Science Process (TDSP) enables best practices across disciplines. Next, you will discover the workflow of the Azure Machine Learning Service and how it can be leveraged on your project. After this, you will review how to create a pipeline for your data preparation, model training, and model registration. Finally, you will explore the infrastructure approaches that can be leveraged for machine learning and how those approaches are supported on Azure. When you are finished with this course, you will possess the skills that will be needed to start a data science project on Azure and the tools that will increase your ability to gain those actionable insights.

Table of contents
  1. Course Overview
  2. Understanding the Azure Machine Learning Workflow
  3. Working with the Azure Machine Learning Workflow
  4. Understanding Data Ingestion Strategies
  5. Understanding Azure Machine Learning Infrastructure

Operationalizing Microsoft Azure AI Solutions

by Ifedayo Bamikole

Dec 16, 2019 / 45m

45m

Start Course
Description

Machine learning and the Azure Artificial Intelligence (AI) platform allow you to use existing data to forecast future behaviors, outcomes, and trends. In this course, Operationalizing Microsoft Azure AI Solutions, you’ll focus on what it means to operationalize AI in Microsoft Azure. First, you’ll go through the lifecycle of an AI model, creating the model in Azure Databricks and Azure Machine Learning Services. Next, you’ll discover how to deploy the model into Azure’s version control tool, Azure Devops, and containerize it such that it can be used by the end user. Finally, you’ll explore how to identify integration points with other Microsoft services and the containers used in Azure - Azure Container Instances and Azure Kubernetes Services. When you’re finished with this course, you’ll have the skills and knowledge of Microsoft Azure needed to devise a strategy for managing version control of an AI solution.

Table of contents
  1. Course Overview
  2. Assembling Appropriate Tools and Technologies
  3. Designing a Version Control Strategy for a Microsoft Azure AI Solution

Advanced

In this section, you will learn how to build out a network of data ingestion sources using Intelligent Edge, as well as how to secure the data and operations of your AI solution.

Designing an Intelligent Edge in Microsoft Azure

by Jared Rhodes

Jul 16, 2019 / 2h 30m

2h 30m

Start Course
Description

Cloud computing has moved more and more out of the cloud and onto the edge. In this course, Designing an Intelligent Edge in Microsoft Azure, you will learn foundational knowledge of edge computing, its intersection with AI, and how to utilize both with Microsoft Azure. First, you will learn the concepts of edge computing. Next, you will discover how to create an edge solution utilizing Azure Stack, Azure Data Box Edge, and Azure IoT Edge. Finally, you will explore how to utilize off the shelf AI and build your own for Azure IoT Edge. When you are finished with this course, you will have the skills and knowledge of AI on the edge needed to architect your next edge solution. Software required: Microsoft Azure, .NET

Table of contents
  1. Course Overview
  2. Designing Solutions That Incorporate AI Pipeline Components on Edge Devices
  3. Identify Appropriate Tools for the Solution
  4. Determine When to Use Built-in Components
  5. Determine When to Build Custom Components

Using Microsoft Azure Security Tools to Protect AI Solutions

by Sahil Malik

Jun 26, 2019 / 1h 45m

1h 45m

Start Course
Description

Securing AI solutions is of paramount importance; not only can AI solutions be hacked, they can be used to hack other systems. In this course, Using Microsoft Azure Security Tools to Protect AI Solutions, you'll explore how AI and security introduce new challenges that we conventionally did not have to worry about. You'll then learn about the various facilities Azure offers that work with Azure ML workspace in helping you secure your AI solutions. Finally, you'll review the various facets you need to consider and the various processes and artifacts you must secure. By the end of this course, you will have a clear vision and understand the various building blocks required to secure Azure AI solutions.

Table of contents
  1. Course Overview
  2. Exploring Security for Azure Machine Learning
  3. AI and Security
  4. Securing Data in Transit and at Rest
  5. Differentiating AI and Platform as a Service Considerations
  6. Wrapping Up