Microsoft Azure AI Engineer (AI-100)

Paths

Microsoft Azure AI Engineer (AI-100)

Authors: James Bannan, Brian Harrison, David Tucker, Jared Rhodes, Tim Warner, Sahil Malik, Ifedayo Bamikole, Ranjan Relan

Microsoft Azure offers a spread of services designed to work together to enable rapid development of high-performance AI solutions. This skill teaches how these Azure services... Read more

What you will learn

  • Ingest, transform, and prepare data for AI solutions
  • Design and implement end-to-end AI solutions on Microsoft Azure
  • Monitor and optimize AI solutions deployed on Microsoft Azure
  • Secure AI solutions on Microsoft Azure

Pre-requisites

This path is intended for learners who are familiar with common AI workflows and concepts, but who do not have experience applying these concepts using Microsoft Azure services.

Designing AI Solutions on Microsoft Azure

This section of the path introduces the relevant services of Microsoft Azure that apply to AI solution design, implementation, and management. In particular, specific attention is given to services that impact how machine learning, security, and Internet-of-Things are addressed in your Azure-enabled AI solution.

Choosing the Appropriate Microsoft Azure Services and Features

by James Bannan

Dec 19, 2019 / 2h 5s

2h 5s

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 28m

1h 28m

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

Designing Machine Learning Solutions on Microsoft Azure

by David Tucker

Dec 6, 2019 / 1h 40m

1h 40m

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

Designing an Intelligent Edge in Microsoft Azure

by Jared Rhodes

Apr 3, 2020 / 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

Developing AI Solutions on MIcrosoft Azure

With the design factors in place, this section of the skill addresses the implementation of the solution design. Specific attention is given to the building of machine learning models, chatbot user experiences, search solutions, and intelligent edge device networks.

Microsoft Azure AI Engineer: Developing ML Pipelines in Microsoft Azure

by Tim Warner

Dec 10, 2019 / 2h 30m

2h 30m

Start Course
Description

At the core of being a Microsoft Azure AI engineer rests the need for effective collaboration. In this course, Microsoft Azure AI Engineer: Developing ML Pipelines in Microsoft Azure, you will learn how to develop, deploy, and monitor repeatable, high-quality machine learning models with the Microsoft Azure Machine Learning service. First, you will understand how to create no-code machine learning pipelines using the Azure ML service visual designer. Next, you will explore how to train ML models using Python, Jupyter notebooks, and the Microsoft Azure Machine Learning workspace. Finally, you will discover how to monitor your Azure Machine Learning environments from the perspective of the data scientist and data engineer. When you are finished with this course, you will have a foundational knowledge of the Microsoft Azure Machine Learning service that will help you as you move forward in the Microsoft Azure AI engineer job role.

Table of contents
  1. Course Overview
  2. Understanding Machine Learning Workspaces
  3. Understanding Azure ML Pipelines
  4. Managing Machine Learning Workspaces
  5. Implementing AI Pipelines
  6. Managing Experiments
  7. Managing Data Flow and Logging

Developing AI Models in Microsoft Azure

by Sahil Malik

May 31, 2019 / 1h 31m

1h 31m

Start Course
Description

AI is all around us, and it is no longer just the work of scientists. In this course, Developing AI Models in Microsoft Azure, you will learn the ins and outs of Azure Machine Learning Service. You'll start with the basics, and learn how to set up your development environment with a demo walking through each important step. With your development environment set up, you'll examine how to use the facilities of the Azure machine learning workspace, interact with it via VSCode and Jupyter notebooks using the Azure ML SDK, how to provision remote compute, and how to deploy a model to the various options, such as docker image, ACI, or AKS. By the end of this course, you will have the necessary skills to tackle any enterprise class custom AI problem in the Microsoft Azure ecosystem.

Table of contents
  1. Course Overview
  2. Approaching AI, ML Studio, and Machine Learning Service
  3. Setting up Your Dev Environment
  4. Training Your Model
  5. Deploying Your Model
  6. Wrapping Up

Implementing a Microsoft Azure AI Bot Framework Solution

by Brian Harrison

May 21, 2020 / 2h 26m

2h 26m

Start Course
Description

When building your bot framework solution in Azure, you will need to understand how to connect it to many different AI and other data-related services as well as how to deploy it into a productionized environment. In this course, Implementing a Microsoft Azure AI Bot Framework Solution, you will gain the ability to develop and deploy a Bot Framework solution. First, you will learn how to connect all of the necessary services for your Bot Framework solution. Next, you will discover how to test the connectivity of those services as well as how to validate the output of the Bot Framework solutions activities. Finally, you will explore how to productionize your Bot Framework solution so that it can be deployed in Azure and you can feel secure about that deployment. When you’re finished with this course, you will have the skills and knowledge of Azure AI Bot Framework needed to develop, test, connect, and eventually deploy your own Bot Framework solution using any of the available connected services that Azure provides.

Table of contents
  1. Course Overview
  2. Setting up the Prerequisite Components and Input Datasets for Consuming Bot Framework
  3. Connecting Pipeline Components
  4. API Output
  5. Productionizing Your Bot

Developing Microsoft Azure Intelligent Edge Solutions

by Jared Rhodes

Sep 12, 2019 / 2h 47m

2h 47m

Start Course
Description

Over time, what was once simply Internet of Things solutions has evolved into Edge solutions. In this course, Developing an Intelligent Edge in Microsoft Azure, you will learn foundational knowledge of edge computing, how it interacts with data and messaging systems, and how to utilize both with Microsoft Azure. First, you will learn the concepts of edge and internet of things computing. Next, you will discover how to process streaming data on hot, warm, and cold paths. Finally, you will explore how real-time and batch processing can be utilized in an edge solution. When you are finished with this course, you will have the skills and knowledge of edge and internet of things in Azure needed to architect your next edge solution. Software required: Microsoft Azure, .NET

Table of contents
  1. Course Overview
  2. Discussing Azure Iot Architecture
  3. Connecting to Iot Hub Data Streams
  4. Implementing Hot, Warm, and Cold Data Streams
  5. Creating Real-time and Batch Processes

Implementing a Microsoft Azure Search Solution

by Sahil Malik

Sep 5, 2019 / 1h 19m

1h 19m

Start Course
Description

Search is so easy for users to use, and they almost expect it in every application. On the other hand, it can be so difficult to build; that simple text box in your application can mean so much work if you have to do it all by yourself. This course, Implementing a Microsoft Azure Search Solution, will help you solve the search dilemma using Azure Search - a cloud-hosted search-as-a-service platform that helps you build compelling search solutions with your data. This powerful capability is not only easy to use, but paired with Cognitive Skills it can help you unlock amazing insights in your unstructured data. At the end of this course, you will feel comfortable with the basic usage of Azure search, plugging in Cognitive Skills in the search pipeline, and administering Azure Search.

Table of contents
  1. Course Overview
  2. Understanding Azure Search
  3. Leveraging Cognitive Search
  4. Administering and Managing

Operationalizing AI Solutions on Microsoft Azure

With an AI solution implemented, this section of the skill turns to AI operations on Microsoft Azure. Deployment, management, and security of the solution are addressed, as well as monitoring and optimization.

Microsoft Azure AI Engineer: Deploying AI Solutions in Microsoft Azure

by James Bannan

Dec 13, 2019 / 2h 20m

2h 20m

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. In this course, Microsoft Azure AI Engineer: Deploying AI Solutions in Microsoft Azure, you’ll gain an understanding of the nature of AI solutions in order to be able to recommend the right services and configuration options which will enable the work of data scientists and ensure that business investment in AI is both fit-for-purpose and well maintained. First, you’ll look at the core nature of AI solutions and how these are represented by solution architectures in Microsoft Azure. Then, you’ll work on the different scenarios to provide the computing services which data scientists need to build, train, and deploy AI models, including understanding the operational frameworks of monitoring, security, and zero-trust architecture. Finally, you’ll bring everything together by deploying an AI model in a container-based solution using IoT Edge. By the end of this course, you’ll have a thorough understanding of how AI solutions in Microsoft Azure are put together, and how you, as an AI Engineer, can influence the architecture, configuration, and deployment of these solutions to support a wide range of business and technical outcomes.

Table of contents
  1. Course Overview
  2. Deploying Azure AI Solutions
  3. Understanding Continuous Monitoring
  4. Deploying Container Environments
  5. Deploying IoT Devices
  6. Managing Security for AI Solutions

Using Microsoft Azure Security Tools to Protect AI Solutions

by Sahil Malik

May 26, 2020 / 1h 55m

1h 55m

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

Operationalizing Microsoft Azure AI Solutions

by Ifedayo Bamikole

Dec 16, 2019 / 44m

44m

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

Managing Microsoft Azure AI Solutions

by Sahil Malik

May 26, 2020 / 1h 38m

1h 38m

Start Course
Description

Companies and Governments across the globe are pouring billions of dollars into AI. The projects are getting ever more interesting and complex, and it is therefore natural to conclude that these projects need management.

In this course, Managing Microsoft Azure AI Solutions, I assert that an AI project is like any other software project, and the need to manage it with good software practices is more, not less, important. With demos, you'll learn how you can use concepts such as Azure CLI, ML SDK, and ML Ops to fully automate your end to end process. You'll also explore how you can set up an Azure DevOps pipeline to go from experiment to a service. But the fun doesn't end there; you'll then discover how to deploy your model as an AKS cluster and enable data monitoring and collection in production, so you can use that data in numerous ways to analyze it or feed it back into your model for subsequent improvement.

By the end of this course, you'll have an in-depth understanding of how to manage your AI projects like a proper software project. Concepts such as ML Ops and Pipelines will be second nature to you, and you'll be a pro at collecting and monitoring your production AI solutions.

Table of contents
  1. Course Overview
  2. Managing Models in Azure Machine Learning Service
  3. Registering Your Model
  4. Registering and Deploying Your Image
  5. Monitoring and Improving Data and Models
  6. Wrapping Up

Optimizing Microsoft Azure AI Solutions

by Ranjan Relan

Dec 19, 2019 / 1h 25m

1h 25m

Start Course
Description

Cloud-based platform Microsoft Azure has multiple AI services which could be used to train your model for big data sets as well as to deploy your model as a web service. In this course, Optimizing Microsoft Azure AI Solutions, you will learn the foundational knowledge of how to train your machine learning models using Azure's services such AzureML Compute Cluster, Azure HDInsight, Azure Databricks, and Azure Data Science Virtual Machine. Next, you will discover how to optimize your storage by using Azure Premium blob storage service and data formats such as Pickle and Parquet. Finally, you will explore how to scale your machine learning models and manage end-to-end machine learning life cycle using the principle of MLOps. When you’re finished with this course, you will have the skills and knowledge of Mircosoft Azure's core AI services needed to design, deploy, and optimize your model.

Table of contents
  1. Course Overview
  2. Optimizing Core Services
  3. Optimizing Storage and Logging
  4. Optimizing Deployments and Operations