Practical agentic AI
AI for Automation: Working with AI Agents
Move beyond one-response prompting and learn how to work effectively with AI agents that can plan, inspect information, use tools, create and revise outputs, and complete multi-step assignments under human direction. The course includes an overview of n8n without making workflow development its main focus.
Build practical capability in AI for Automation: Working with AI Agents.
Move beyond one-response prompting and learn how to work effectively with AI agents that can plan, inspect information, use tools, create and revise outputs, and complete multi-step assignments under human direction. The course includes an overview of n8n without making workflow development its main focus.
Distinguish AI assistants, automated workflows, and AI agents and select appropriate workplace applications.
Delegate multi-step assignments to AI agents using clear goals, context, constraints, files, tools, and completion criteria.
Guide, monitor, review, and redirect agent work while maintaining appropriate human control.
Apply privacy, security, validation, approvals, auditability, and governance to agent-assisted work.
Course overview
A structured, application-focused AI program. Open each topic to review its practical coverage.
Module 1: From AI Assistants to AI Agents
- AI Assistants vs AI Workflows vs AI Agents
- From Automation to Agentic AI
Module 2: How AI Agents Work
- The Components of an AI Agent
- Inputs, Memory, Tools, and Outputs
Module 3: Giving an Agent a Clear Assignment
- Defining the Goal and Expected Outcome
- Providing Context, Source Files, and Reference Material
Module 4: Working with Agents on Multi-Step Tasks
- Breaking Complex Work into Stages
- Allowing the Agent to Plan and Sequence Work
Module 5: Agents Using Tools and Taking Actions
- Reading and Creating Files
- Searching and Gathering Information
Module 6: Reviewing and Validating Agent Output
- Checking Completeness Against the Original Goal
- Identifying Hallucinations, Errors, and Unsupported Claims
Module 7: Reusable Agent Workflows
- Building Reusable Instructions for Recurring Tasks
- Creating Templates, Checklists, and Reference Packages
Module 8: Overview of n8n and Workflow Platforms
- What n8n Is and Where It Fits
- Nodes, Connections, Triggers, and Actions
Module 9: Responsible Agent Use
- Privacy Considerations and Confidential Information
- Security Basics and Permission Management
Module 10: Hands-On Agent Assignment
- Define a Useful Multi-Step Workplace Task
- Prepare the Instructions, Files, Constraints, and Completion Criteria
Full course outline
This restructured outline uses the strongest applicable topics from the legacy AI programs while removing unnecessary overlap.
View full outline
- Module 1: From AI Assistants to AI Agents
- AI Assistants vs AI Workflows vs AI Agents
- From Automation to Agentic AI
- Thinking, Planning, and Acting
- Current Capabilities and Limitations
- Real-world Business Applications
- Module 2: How AI Agents Work
- The Components of an AI Agent
- Inputs, Memory, Tools, and Outputs
- Large Language Models and Context Windows
- Single-Agent vs Multi-Agent Systems
- Human-in-the-Loop Concepts
- Module 3: Giving an Agent a Clear Assignment
- Defining the Goal and Expected Outcome
- Providing Context, Source Files, and Reference Material
- Setting Scope, Constraints, and Non-Negotiable Requirements
- Defining Completion Criteria and Quality Standards
- Using Effective System and Task Instructions
- Module 4: Working with Agents on Multi-Step Tasks
- Breaking Complex Work into Stages
- Allowing the Agent to Plan and Sequence Work
- Maintaining Context Across a Long Assignment
- Reviewing Progress and Providing Corrections
- Knowing When to Continue, Redirect, or Stop the Agent
- Module 5: Agents Using Tools and Taking Actions
- Reading and Creating Files
- Searching and Gathering Information
- Working with Documents, Data, and Applications
- Calling APIs and Connecting External Services
- Triggering Automated Actions
- Understanding Permissions and Approval Boundaries
- Module 6: Reviewing and Validating Agent Output
- Checking Completeness Against the Original Goal
- Identifying Hallucinations, Errors, and Unsupported Claims
- Validating Files, Calculations, Links, and Actions
- Requesting Revisions Without Losing Good Work
- Creating Review Checklists and Acceptance Tests
- Module 7: Reusable Agent Workflows
- Building Reusable Instructions for Recurring Tasks
- Creating Templates, Checklists, and Reference Packages
- Using Memory and Knowledge Sources Appropriately
- Coordinating Multiple Agents or Specialized Roles
- Documenting the Human and Agent Responsibilities
- Module 8: Overview of n8n and Workflow Platforms
- What n8n Is and Where It Fits
- Nodes, Connections, Triggers, and Actions
- AI Agent Nodes, Tools, Memory, and Structured Outputs
- When to Use a Workflow Platform Instead of an Interactive Agent
- Examples of Connected Business Automations
- Module 9: Responsible Agent Use
- Privacy Considerations and Confidential Information
- Security Basics and Permission Management
- Human Oversight and Approval Workflows
- Auditability, Logs, and Traceability
- AI Governance Fundamentals
- Module 10: Hands-On Agent Assignment
- Define a Useful Multi-Step Workplace Task
- Prepare the Instructions, Files, Constraints, and Completion Criteria
- Guide an AI Agent Through Planning and Execution
- Review, Test, and Revise the Agent's Output
- Document the Final Workflow and Human Controls