Designing and Implementing an Azure AI solution (AI-100 exam is designed for candidates having subject matter expertise and knowledge in using cognitive services, machine learning, and knowledge mining to architect and implement Microsoft AI solutions. In addition, this also involves natural language processing, speech, computer vision, and conversational AI. This exam will help the candidates in becoming Microsoft Certified: Azure AI Engineer Associate.
Microsoft AI-100 exam is getting retired on June 30, 2021. A new replacement exam Designing and Implementing a Microsoft Azure AI Solution Beta (AI-102) is available.
However, an Azure AI Engineer should know how to analyze requirements for AI solutions with recommending the appropriate tools and technologies, and designing and implementing AI solutions that meet scalability and performance requirements. And, they must have knowledge to translate the vision from solution architects. However, Azure AI Engineers work with data scientists, data engineers, IoT specialists, and software developers to build complete end-to-end solutions.
Knowledge requirement for the exam
Candidates applying for this exam should have:
- Knowledge and experience in designing and implementing AI apps and agents that use Microsoft Azure Cognitive Services.
- Knowledge about Azure Bot Service, Azure Cognitive Search, and data storage in Azure.
- Ability to suggest solutions that use open source technologies with understanding the components that make up the Azure AI portfolio and the available data storage options.
- Understanding when a custom API should be developed to meet specific requirements.
Microsoft provides exam objectives that help the candidates to understand and know about the concepts before preparation. Moreover, these exam concepts are provided with sections and subsections to make you learn about it in depth. For AI-100, The Microsoft includes training resources that provide a learning path to help you during the studying time. However, the basic concepts include:
- Analyzing the requirements for solution
- Designing AI solutions
- Implementing and monitoring the AI solutions
Microsoft provides candidates access to the learning path that helps them to understand the concepts in a step by step format. However, these learning paths include modules that help candidates to enhance their skills and knowledge in:
- Evaluating text with Azure cognitive language services
- Processing and Translating speech with azure cognitive speech services
- Creating intelligent bots with the Azure Bot Service
- Processing and classifying images with the Azure Cognitive Vision Service
Designing and Implementing an Azure AI Solution (AI-100) exam consists of 62 questions. To complete the exam, candidates will get 220 minutes. The types of questions that candidates can face in the exam include scenario-based single answer questions, multiple-choice questions, arrange in the correct sequence type questions, drag & drop questions, mark review, drag, and drop, etc. However, to pass the exam, a candidate has to score a minimum of 700 or more. To apply for the exam, the examination fee is $165 USD including taxes. And, candidates can give the exam in English, Japanese, Chinese (Simplified) and Korean languages.
Microsoft Azure AI-100 exam measures the ability to perform technical tasks for analyzing solution requirements with designing solutions and integrating AI models into solutions. Candidates planning to take the AI-100 exam should also have a strong understanding in Machine Learning and implementing Microsoft AI solutions. Candidates can schedule their exam with the
Course Outline for Microsoft AI-100 Exam
Topic 1: Analyze solution requirements
1.1 Recommending Azure Cognitive Services APIs to meet business requirements
Microsoft Documentation: Azure Cognitive Services
- selecting the processing architecture for a solution (Microsoft Documentation: Machine Learning Products)
- Choosing the appropriate data processing technologies (Microsoft Documentation: Choosing Data Store)
- selecting the appropriate AI models and services
- identifying components and technologies required to connect service endpoints (Microsoft Documentation: Components of REST)
- identify automation requirements (Microsoft Documentation: Azure Automation)
1.2 Mapping security requirements for tools, technologies, and processes
- identifying processes and regulations needed to conform to data privacy, protection, and regulatory requirements (Microsoft Documentation: Regulatory compliance)
- identify which users and groups have access to information and interfaces (Microsoft Documentation: Permission, user and groups in Azure)
- locating appropriate tools for a solution (Microsoft Documentation: Azure Security)
- identify auditing requirements (Microsoft Documentation: Azure Auditing)
1.3 Selecting the software, services, and storage required to support a solution
- identifying appropriate services and tools for a solution (Microsoft Documentation: Azure Machine Learning, Microsoft Cognitive Service)
- locating integration points with other Microsoft services (Microsoft Documentation: Azure Event Grid, Azure Event Hubs)
- identify storage required to store logging, bot state data, and Azure Cognitive Services output (Microsoft Documentation: Data Store )
Topic 2: Design AI solutions
2.1 Designing solutions that includes one or more pipelines
- defining an AI application workflow process (Microsoft Documentation: Azure Machine Learning Pipeline)
- design a strategy for ingesting and egress data
- designing the integration point between multiple workflows and pipelines
- outlining pipelines that use AI apps (Microsoft Documentation: Managing Web Services )
- plotting pipelines that call Azure Machine Learning models (Microsoft Documentation: Creating web services endpoints)
- selecting an AI solution that meets cost constraints
2.2 Designing solutions that uses Cognitive Services
- Planning solutions that use vision, speech, language, knowledge, search, and anomaly detection APIs (Microsoft Documentation: Recognize speech from Microphone, Get intent with REST API, Visualise Anomalies )
2.3 Design solutions that implement the Microsoft Bot Framework
- integrating bots and AI solutions
- outlining bot services that use Language Understanding (LUIS) (Microsoft Documentation: Using Web App Bot)
- modeling bots that integrate with channels (Microsoft Documentation: Connect Bot to channels)
- integrating bots with Azure app services and Azure Application Insights (Microsoft Documentation: Create Azure Web App Bot)
2.4 Designing the compute infrastructure to support a solution
- identifying whether to create a GPU, FPGA or CPU-based solution (Microsoft Documentation: FPGA)
- identify if to use a cloud-based, on-premises, or hybrid compute infrastructure
- selecting a compute solution that meets cost constraints (Microsoft Documentation: Choosing Azure Compute Service)
2.5 Model data governance, compliance, integrity, and security
- defining how users and applications will authenticate to AI services (Microsoft Documentation: Authenticate request to Azure Cognitive Service)
- designing a content moderation strategy for data usage within an AI solution (Microsoft Documentation: Azure Content Moderator)
- ensuring that data adhere to compliance requirements defined by your organization (Microsoft Documentation: Get compliance data of Azure Resources, Microsoft Compliance Manager)
- ensuring appropriate governance of data (Microsoft Documentation: Azure Governance)
- designing strategies to ensure that the solution meets data privacy regulations and industry standards (Microsoft Documentation: Data collection, retention, and storage in application )
Topic 3: Implement and monitor AI solutions
3.1 Implementing an AI workflow
- developing AI pipelines (Microsoft Documentation: Azure Machine Learning Pipelines)
- managing the flow of data through the solution components (Microsoft Documentation: Azure IoT, Advanced Analytics Architecture)
- implementing data logging processes (Microsoft Documentation: Diagnostic Logging)
- defining and construct interfaces for custom AI services (Microsoft Documentation: Configure Bing Custom Search)
- creating solution endpoints (Microsoft Documentation: Using Azure Events Hub)
- developing streaming solutions (Microsoft Documentation: Azure Stream Analytics Solutions)
3.2 Integrating AI services and solution components
- configuring prerequisite components and input datasets to allow the consumption of Azure Cognitive Services APIs (Microsoft Documentation: Building training data set)
- configuring integration with Azure Cognitive Services (Microsoft Documentation: Configuring apps to expose web APIs)
- configure prerequisite components to allow connectivity to the Microsoft Bot Framework (Microsoft Documentation: Create Bot with Azure Bot Service)
- implementing Azure Cognitive Search in a solution (Microsoft Documentation: Search Web using Bing Web Search REST APIs and C#)
3.3 Monitoring and evaluating the AI environment
- identifying the differences between KPIs reported metrics and root causes of the differences (Microsoft Documentation: Create custom KPI dashboard)
- locating the differences between expected and actual workflow throughput (Microsoft Documentation: Scalability and Performance, Monitor and collect data from ML web services endpoints)
- maintaining an AI solution for continuous improvement (Microsoft Documentation: Create CI/CD pipelines using Azure pipelines, docker and Kubernetes)
- monitoring AI components for availability (Microsoft Documentation: Application insights telemetry data model, Collect Azure Platform logs in logs analytics workspace in Azure Monitor)
- recommending changes to an AI solution based on performance data
Microsoft provides exam policies to help the candidates to plan and manage a positive outcome. Microsoft Certification exam policies give candidates access to all the exam-related details, accompanying the before and after exam procedures. These exam policies are the inclusion of certain rules that need to be followed during the exam time or at testing centers.
For more Queries Visit: Designing and Implementing an Azure AI Solution (AI-100) Exam FAQs
Preparation Guide for Microsoft Azure AI-100 Exam
Microsoft Learning Platform
The learning resource that will be beneficial during the exam preparation is the Microsoft learning platform. However, make sure to go through the official website of Microsoft. For the AI-100 exam, it would be best to first go through the Microsoft official website to get authentic information about the exam. You can easily locate the AI-100 page where you can just go through all the necessary information about the AI-100 exam.
After that, you can move on to Microsoft documentation where you can easily understand the Microsoft AI solutions and Machine learning concepts. Moreover, you also get to know the different scales of different Azure services. Microsoft Docs consists of modules that will help you gain a lot of knowledge about AI and the different services in a sequence.
Microsoft has given an advantage to candidates providing instructor-led training. This training will help students to gain knowledge designing Azure AI solutions by building customer support chat Bot using artificial intelligence from the Microsoft Azure platform including language understanding and pre-built AI functionality in the Azure Cognitive Services. However, this training is designed for Cloud Solution Architects, Azure artificial intelligence designers, and AI developers.
Online Study Groups
One thing that will be beneficial during the exam preparation time is to join study groups. As these groups will help you to stay connected with the other people who are on the same pathway as yours. Moreover, here you can start any discussion about the issue related to the exam or any query. By doing so, you will get the best possible answer to your query.
Those who are dedicated to passing the exam know the importance of books during the time of preparation. However, while studying for the exam books can be really helpful to understand the core of the topics. Candidates can take the books available in the market that will help in studying for the AI-100 exam.
This can be a very essential part that can help you to prepare better for the exam. That is to say, practice tests are important as by assessing yourself with these tests you will know about your weak and strong areas. However, by practicing you will be able to improve your answering skills that will result in saving a lot of time. Moreover, the best way to start doing practice tests is after completing one full topic as this will work as a revision part for you. So, make sure to find the best practice sources.