No-Code Agentic AI: Building Business Automation with n8n

Duration

2 Days

Skills Gained

By the end of this course, participants will be able to:

• Define agentic AI as distinct from other AI techniques

• Describe the components of an AI agent (model, memory, tools, OODA loop)

• Configure best practices for designing and implementing AI agents

• Build basic AI agents for automation and human collaboration in n8n

• Identify key ethical and security risks of agentic AI

Who Can Benefit

This course is designed for:

• Business Analysts & Process Owners

• Operations & Support Teams

• Team Leads & Managers Automating Workflows

• Anyone New to Agentic AI

Prerequisites

Participants should enter this course with:

• Familiarity with basic Generative AI concepts such as prompting

Organizational Objectives

This course assists organizations to:

• Reduce time-to-first-agent for business teams through no-code n8n workflows

• Build a shared understanding of agentic AI capabilities and limits across non-engineering teams

• Lower governance risk through trained recognition of excessive-agency and ethical issues, including PHI exposure

• Standardize human-in-the-loop patterns for agent-driven automation

Software

All attendees must have a modern web browser and an Internet connection.

Module 1 - Introduction to Agentic AI

By the end of this module, you will be able to define agentic AI, differentiate it from other AI techniques, articulate its benefits and challenges, and apply it across collaboration and automation use cases.

• Defining agentic AI and its key characteristics

• Differentiating agentic AI from other techniques

• Benefits and challenges of agentic AI

• Agentic AI for collaboration and automation

• Real-world applications of agentic AI

• Hands-on Lab: Stand up your agent platform and build a small introductory agent that answers common member questions about plan benefits, to verify the toolchain end-to-end.

Module 2 - Components of an Agentic AI

By the end of this module, you will be able to select the right LLM for an agent task, recognize the role memory plays, distinguish working, long-term, and episodic memory, and apply the OODA loop to agentic decision-making.

• Models — what is in an LLM?

• Selecting the right model for the task

• The role of memory in agentic AI

• Types of memory — session-based (Simple Memory node), persistent (Postgres/Redis), and episodic

• The OODA loop in agentic AI — observe, orient, decide, act

• Hands-on Lab: Select an appropriate model and add working and long-term memory to a member-service agent so it remembers a caller’s plan type across a multi-turn conversation.

Module 3 - Tool Use and MCP Servers

By the end of this module, you will be able to distinguish observation tools from action tools, apply direct and indirect tool-use methods, force tool use when needed, and integrate MCP servers into your agent.

• Observation vs. action tools

• Methods of tool use — direct, indirect

• Forcing tool use in agents

• Overview of MCP servers

• n8n as both MCP client and MCP server: consuming external tools vs. exposing your own workflows as tools

• Common MCP servers and their use cases

• Hands-on Lab: Add a claims-status lookup tool to your agent and integrate one MCP server so it can check a sample prior authorization’s status.

Module 4 - Building Agentic AI Applications

By the end of this module, you will be able to design an agentic AI application end-to-end, decompose its tasks across single or multiple agents, evaluate agent behaviour through testing, handle edge cases and failures gracefully, and apply best practices for building resilient agents.

• Designing Agentic AI Applications

• Task Decomposition

• Multi-Agent Systems (AI Agent Tool node for single-canvas orchestration)

• Testing and Evaluating Agents

• Handling Edge Cases and Failures

• Best Practices for Creating Agents

• Hands-on Lab: Design and build a resilient prior-authorization intake agent that survives at least one deliberate edge case, such as an incomplete claim submission.

Module 5 - Designing Collaborative Agents

By the end of this module, you will be able to choose between automation and human collaboration for an agent task, design communication and coordination between humans and agents, build trust through transparency, and design human-agent interaction patterns.

• Automation vs. human collaboration

• Communication and coordination mechanisms

• Building trust between humans and agents

• Case studies of collaborative agents

• Designing for human-agent interaction

• Hands-on Lab: Build a human-in-the-loop workflow for a coverage-denial appeal agent, and debug an agent failure with a claims specialist as the recovery path.

Module 6 - Ethics and Security of Agentic AI

By the end of this module, you will be able to identify ethical challenges of autonomous agents, apply fairness, transparency, and accountability principles, recognize key security threats and excessive-agency risks, and mitigate misinformation and undesirable outcomes.

• Ethical challenges of autonomous agents

• Fairness, transparency, and accountability

• Key security threats and excessive agency risks

• Mitigating misinformation and undesirable outcomes

• Building a culture of responsible AI

• Hands-on Lab: Audit a claims-processing agent design for PHI exposure, excessive-agency, and fairness risks, and propose mitigations grounded in Responsible AI principles.