AB-900: Copilot & Agent Admin Course +360 Practice Questions

Explore the Microsoft 365 ecosystem as a cloud-based productivity platform unifying apps, security, identity, and devices across four pillars: productivity apps, security and compliance, identity and access, and device management.

Explore the Microsoft 365 admin center to manage users, licenses, and group-based licensing with security features like MFA. Navigate Exchange, team, and SharePoint admin centers and monitor governance and roles.

Explain RBAC in Microsoft 365, detailing how built-in roles grant specific permissions, enforce the principle of least privilege, and assign global, user, exchange, and helpdesk admin roles.

Learn how to protect and govern data in Microsoft 365 using Microsoft Purview, a unified platform that discovers, classifies, labels, and protects sensitive information.

Discover how Microsoft Purview links Compliance Manager, AI data discovery (DSPM), and eDiscovery to map sensitive data, boost compliance scores, detect oversharing, govern Copilot access, and export legal evidence.

Microsoft 365 Copilot and ai agents, learn their architecture and licensing, plan deployments, and master governance, administration, and ongoing optimization across the four learning modules.

Explore Microsoft 365 Copilot and agents through concrete use cases, uncover the three core components powering Copilot, security boundaries, licensing models, and deployment planning.

Explore how Microsoft 365 Copilot lives inside the apps you use daily, drafting relevant project updates from your emails, Teams notes, and SharePoint documents while respecting all permission boundaries.

Copilot relies on three components: Microsoft Graph, large language models, and the orchestration service, to securely provide memory, permissioned data access, and safe contextual responses.

See how Copilot integrates into Word, Excel, Outlook, and Teams to draft, analyze, and summarize, and compare Copilot chat with Microsoft 365 Copilot for licensing and data access.

Discover how Microsoft 365 Copilot embeds generative AI into Word, Excel, PowerPoint, Outlook, and Teams with Graph, LLMs, secure orchestration, agents, and licensing basics for deployment readiness.

discover the four-layer copilot architecture—identity, service, data, and intelligence—and learn admin troubleshooting, isolation, business-tuned llms, and scalable, secure deployment within your tenant.

Explore the five Copilot licensing doors from free Microsoft 365 Copilot Chat to per-user add-ons, pay-as-you-go credits, and personal plans, highlighting base licenses and hybrid governance.

Learn when pay-as-you-go copilot billing is advantageous for experimentation, seasonal work, and autonomous agents, and how to set up budgets, guardrails, and Azure–M365 roles to control costs.

Evaluate copilot adoption by tracking enabled vs active users, then use the Viva Insight and admin center dashboards to measure usage patterns, app breakdown, and return on investment.

Create a SharePoint agent with Copilot Studio's readymade integration by selecting up to 20 documents, defining sources, behavior, and startup prompts to ground responses.

Manage agent access by assigning a Microsoft 365 copilot license and configuring environment roles (environment maker, system customizer) with best practices for least privilege and dev-prod separation.

Monitor and manage the agent life cycle with three lenses—permission usage, cross-service dependencies, and trend analysis. Run a monthly review to trim unused permissions.

Master the four pillars of co-pilot administration—create, test and edit, access and permission, and govern and monitor—to keep agents in your tenant secure, compliant, and aligned with business needs.

Explore AI, ML, neural networks, and deep learning from a high-level perspective, and learn a developer roadmap to build intelligent apps with prompts and AI APIs.

Explore how ai-driven applications differ from traditional non-ai apps by enabling natural language input, unstructured data processing, and continuous learning that evolves tools like chatbots, recommendation systems, and autonomous vehicles.

Explore three AI types—narrow AI, artificial general intelligence (AGI), and super AI—through real-world tasks like Siri, Alexa, and Netflix recommendations, and discuss the current gap to AGI.

Explore ai workloads from machine learning and computer vision to natural language processing and generative ai, with document intelligence, anomaly detection, and knowledge mining powering cross-industry use cases.

Learn how machine learning learns from data to find patterns and build models for predictions and decisions as a subset of artificial intelligence, guided by supervised, unsupervised, and reinforcement learning.

Explore common machine learning algorithms like linear and logistic regression, Naive Bayes, decision trees, SVM, neural networks, and k-means clustering, and learn to select and train a model.

Understand how a machine learning model is created from training data and algorithms to predict outcomes, and learn to evaluate it with a training and test data split.

Evaluate ml models using accuracy, precision, recall, f1, mse, and rmse across the workflow from problem definition to deployment and monitoring; explore careers like data analyst and ml engineer.

Train a supervised machine learning model in Python on a diabetes dataset using logistic regression, split data, evaluate with a confusion matrix and accuracy, and deploy via a web app.

Explore fundamentals of artificial neural networks and deep learning, from perceptron basics to multi-layer networks that learn features automatically, using activation functions, backpropagation, and GPUs for unstructured data like images.

Explore the four main neural network architectures—fully connected, convolutional (cnn), recurrent (rnn), and autoencoders—and their use in vision, sequential data, and content generation.