• Solution Architects and Enterprise Architects
• Dynamics 365 architects and senior consultants
• Power Platform solution architects
• Microsoft 365 and Copilot architects
• AI transformation leads
• Technical leads responsible for agentic AI strategy
• Consultants preparing for Microsoft Certified: Agentic AI Business Solutions Architect

Prerequisites

Learners should have experience with:

• Microsoft cloud and business applications
• Dynamics 365, Power Platform, or Microsoft 365 architecture
• Solution design and enterprise integration concepts
• Data, security, governance, and ALM fundamentals
• Basic generative AI and Copilot concepts

Module 1: Introduction to Agentic AI Business Architecture

Topic 1: Understanding Agentic AI in Business Solutions

• Explains how agentic AI differs from traditional automation, chatbots, and workflow tools.
• Introduces agents as goal-oriented systems that can reason, take action, and interact with business data.
• Connects agentic AI to business transformation, productivity, customer experience, and operational redesign.

Topic 2: Microsoft Agentic AI Solution Landscape

• Reviews the roles of Microsoft Copilot, Microsoft 365 Copilot, Copilot Studio, Microsoft Foundry, Dynamics 365, and Power Platform.
• Explains how Microsoft platforms support prebuilt agents, custom agents, AI prompts, business apps, and enterprise data grounding.
• Helps learners understand when to use Microsoft-native agents, custom agents, copilots, or extended business applications.

Topic 3: Architect Responsibilities for AI-First Solutions

• Defines the role of the AI-first solution architect in strategy, design, governance, adoption, and implementation.
• Emphasizes architectural judgment, business value alignment, security, and lifecycle ownership.
• Connects AB-100 responsibilities to enterprise architecture, solution validation, stakeholder alignment, and measurable outcomes.

Module 2: Analyze Business Requirements for AI-Powered Solutions

Topic 1: Identifying Business Processes for AI Transformation

• Teaches learners how to identify processes where agents can automate tasks, improve decisions, or reduce manual effort.
• Covers use cases across sales, service, finance, supply chain, operations, HR, and knowledge work.
• Helps architects distinguish between simple automation, assisted AI, agentic workflows, and autonomous agents.

Topic 2: Assessing Agent Use Cases

• Evaluates where agents can support task automation, analytics, decision-making, and process orchestration.
• Reviews criteria such as business impact, complexity, risk, repeatability, data availability, and user adoption.
• Helps learners prioritize use cases that are realistic, measurable, and aligned to enterprise goals.

Topic 3: Reviewing Data Readiness for Grounding

• Covers data accuracy, relevance, timeliness, cleanliness, and availability.
• Explains why grounding data quality directly affects AI response quality, trust, and usefulness.
• Helps learners evaluate Dataverse, Dynamics 365, SharePoint, Microsoft Graph, Fabric, external systems, and knowledge sources.

Module 3: Design the AI Strategy for Business Solutions

Topic 1: AI Adoption and Cloud Adoption Framework Alignment

• Introduces Microsoft’s AI adoption concepts through the Cloud Adoption Framework for Azure.
• Shows how governance, strategy, readiness, adoption, management, and innovation apply to AI-powered solutions.
• Helps architects align AI projects with business value, risk management, and enterprise operating models.

Topic 2: Designing the Agentic AI Roadmap

• Defines how to create a phased roadmap for agents, copilots, AI prompts, custom models, and business app extensions.
• Helps organizations move from pilot projects to scalable enterprise AI capabilities.
• Connects roadmap planning to business capability maps, process redesign, platform readiness, and adoption maturity.

Topic 3: Establishing an AI Center of Excellence

• Explains the purpose of an AI Center of Excellence for governance, reuse, standards, training, and responsible AI.
• Covers roles such as business sponsor, solution architect, data owner, security lead, governance lead, and adoption lead.
• Helps learners design operating models that support repeatable, safe, and scalable AI delivery.

Module 4: Evaluate Build, Buy, Extend, and ROI Decisions

Topic 1: Build, Buy, or Extend AI Components

• Teaches architects how to decide whether to use prebuilt Microsoft AI capabilities, extend Copilot, build custom agents, or develop custom models.
• Reviews trade-offs involving cost, complexity, time-to-value, control, security, maintainability, and business differentiation.
• Helps learners recommend practical AI solution patterns for enterprise scenarios.

Topic 2: ROI and Total Cost of Ownership

• Covers ROI criteria for AI-powered business solutions, including efficiency gains, revenue impact, service improvement, risk reduction, and user productivity.
• Explains total cost of ownership across licensing, consumption, implementation, governance, maintenance, and support.
• Helps learners build a business case that compares expected value against implementation and operating costs.

Topic 3: Model Routing and Cost Optimization

• Introduces model routing as a strategy for sending requests to the most appropriate model based on task, cost, speed, risk, and quality.
• Explains how model choice can affect solution performance, cost, accuracy, and governance.
• Helps learners design AI solutions that balance enterprise needs with responsible cost management.

Module 5: Design Agents and AI Components for Business Solutions

Topic 1: Designing Task Agents, Autonomous Agents, and Prompt Agents

• Explains the differences between task agents, autonomous agents, and prompt-and-response agents.
• Shows how to match agent type to business scenario, risk level, required oversight, and process complexity.
• Helps learners design agent behavior, escalation paths, human-in-the-loop review, and business controls.

Topic 2: Designing Copilot Studio Topics, Actions, and Fallback

• Covers agent topics, prompt actions, fallback behavior, and conversational design in Copilot Studio.
• Explains how topics and actions support structured business processes and user interactions.
• Helps learners design agents that are useful, predictable, secure, and maintainable.

Topic 3: Designing AI in Power Apps and Business Applications

• Shows how AI components can be embedded into Power Apps canvas apps and model-driven apps.
• Reviews business process design patterns that combine user interfaces, workflows, AI prompts, connectors, and Dataverse.
• Helps learners apply Power Platform Well-Architected principles to intelligent application workloads.

Module 6: Architect AI Solutions for Dynamics 365

Topic 1: AI for Dynamics 365 Customer Experience and Service

• Covers Copilot and agent design for Dynamics 365 Sales, Customer Service, Contact Center, Field Service, and Customer Insights scenarios.
• Explains business terms, connectors, customizations, and service channel integration.
• Helps learners design AI-supported customer engagement, case resolution, sales productivity, and service automation.

Topic 2: AI for Dynamics 365 Finance and Supply Chain

• Reviews AI features and agent scenarios for finance, procurement, supply chain, operations, and in-app assistance.
• Covers how finance and operations agent chats can use additional knowledge sources.
• Helps learners design AI-supported business processes for ERP users, operational decision-making, and guided work.

Topic 3: Cross-App Dynamics 365 AI Solution Design

• Teaches how to design AI solutions that span multiple Dynamics 365 applications.
• Covers end-to-end business processes such as lead-to-cash, case-to-resolution, procure-to-pay, and demand-to-deliver.
• Helps learners design integrated AI experiences across CRM, ERP, Power Platform, Microsoft 365, and external systems.

Module 7: Design AI Extensibility with Microsoft Foundry and Microsoft 365

Topic 1: Microsoft Foundry and Custom Models

• Introduces Microsoft Foundry tools, models, and custom AI solution patterns.
• Covers when custom models should be created, fine-tuned, or avoided.
• Helps architects design AI components that meet specialized business, data, compliance, or performance requirements.

Topic 2: Microsoft 365 Copilot and Agent Extensibility

• Explains how Microsoft 365 Copilot can be extended with agents, connectors, Teams, SharePoint, and Microsoft Graph.
• Covers business scenarios for Microsoft 365 agents across productivity, collaboration, knowledge management, and service delivery.
• Helps learners design AI experiences where employees already work.

Topic 3: Model Context Protocol and Agent Interoperability

• Introduces extensibility patterns using Model Context Protocol and agent interoperability concepts.
• Explains how external tools, data sources, apps, and services can be made available to agents.
• Helps learners evaluate interoperability, security, governance, and maintainability risks.

Module 8: Orchestrate Prebuilt Agents, Copilots, and AI Features

Topic 1: Orchestrating Microsoft 365 Copilot for Sales and Service

• Reviews the configuration and business value of Microsoft 365 Copilot for Sales and Microsoft 365 Copilot for Service.
• Explains how these copilots connect Microsoft 365 productivity experiences with CRM and service workflows.
• Helps learners design adoption and integration strategies for seller and service agent productivity.

Topic 2: Orchestrating Power Platform AI Features

• Covers AI Builder, AI prompts, AI hub, Copilot Studio, connectors, Dataverse, and Power Automate integration.
• Explains how Power Platform AI features can support low-code intelligent applications.
• Helps learners choose the right Power Platform AI feature for business requirements.

Topic 3: Orchestrating Knowledge Sources

• Explains how knowledge sources improve the quality and relevance of agent responses.
• Covers sources such as SharePoint, Dataverse, Dynamics 365 records, external systems, websites, and curated knowledge bases.
• Helps learners design knowledge strategies that support grounding, access control, compliance, and user trust.

Module 9: Test and Validate AI-Powered Business Solutions

Topic 1: Agent Testing Strategy

• Covers test planning, success criteria, validation metrics, and expected behavior for agents.
• Explains how to test conversations, prompts, actions, workflows, fallback, escalation, and exception handling.
• Helps learners define quality gates before agent deployment.

Topic 2: Prompt Validation and Evaluation

• Teaches learners how to validate prompt effectiveness, consistency, accuracy, tone, and business alignment.
• Covers prompt libraries, prompt engineering guidelines, and reusable prompt patterns.
• Helps architects design prompt governance that reduces risk and improves repeatability.

Topic 3: End-to-End AI Test Scenarios

• Covers testing AI solutions across multiple Dynamics 365 apps, Power Platform components, Microsoft 365, and external systems.
• Explains how to validate data flow, security, user experience, model behavior, and process outcomes.
• Helps learners create business process–based test scenarios instead of isolated feature tests.

Module 10: Monitor, Tune, and Optimize Agentic AI Solutions

Topic 1: Monitoring Agent Performance

• Covers tools and processes for monitoring agents, usage, errors, satisfaction, resolution rates, and business outcomes.
• Explains how telemetry supports operational visibility and continuous improvement.
• Helps learners define dashboards and metrics for agent reliability and business value.

Topic 2: Interpreting Telemetry and User Feedback

• Teaches learners how to analyze backlog, usage patterns, user feedback, and agent performance data.
• Explains how feedback loops can improve prompts, knowledge sources, actions, and model behavior.
• Helps architects design lifecycle practices that keep AI solutions aligned with changing business needs.

Topic 3: Tuning AI Models and Agent Behavior

• Covers tuning strategies for prompts, grounding data, model selection, response quality, and agent workflows.
• Explains how AI-based tools can help identify issues and recommend improvements.
• Helps learners optimize agent performance while maintaining governance and responsible AI requirements.

Module 11: Design ALM for AI-Powered Solutions

Topic 1: ALM for Copilot Studio Agents

• Covers ALM for Copilot Studio agents, topics, actions, connectors, environments, and dependencies.
• Explains how agents should move through development, test, staging, and production environments.
• Helps learners design release management processes for low-code and pro-code AI components.

Topic 2: ALM for Microsoft Foundry and Custom Models

• Reviews lifecycle management for Foundry agents, custom models, model versions, data assets, and evaluation results.
• Explains how model changes should be tracked, validated, approved, and deployed.
• Helps learners design AI model governance that supports auditability and operational control.

Topic 3: ALM for Dynamics 365 and Power Platform AI

• Covers ALM for AI features embedded in Dynamics 365, Power Apps, Power Automate, Dataverse, and business applications.
• Explains how solution layering, environment strategy, configuration data, and deployment pipelines affect AI workloads.
• Helps learners design ALM processes that support enterprise reliability and repeatable delivery.

Module 12: Responsible AI, Security, Governance, Risk, and Compliance

Topic 1: Security for Agents and AI Workloads

• Covers identity, access control, least privilege, data permissions, grounding data security, and model access.
• Explains how agents can introduce new risks when they access systems, data, tools, and actions.
• Helps learners design secure architectures that prevent unauthorized access or unintended automation.

Topic 2: Responsible AI and Risk Management

• Reviews Microsoft responsible AI principles and how they apply to business solutions.
• Covers fairness, reliability, safety, privacy, security, inclusiveness, transparency, and accountability.
• Helps learners design governance controls for AI risk, review, approval, monitoring, and escalation.

Topic 3: Compliance, Audit, and Data Residency

• Covers data residency, data movement, audit trails, model tuning controls, and regulatory considerations.
• Explains how architects should validate compliance across data, prompts, models, actions, and integrations.
• Helps learners design AI solutions suitable for enterprise, regulated industry, and public sector environments.