Course Outline
Day 1: Workflows and the AI Step
Introduction
- Determining when to automate, when to incorporate an AI step, and when to use an agent
- n8n in context: comparing cloud and self-hosted options and their implications for company data
n8n Fundamentals
- Workflows, nodes, triggers, and executions
- Transferring data between nodes by dragging fields, eliminating the need for syntax entry
- Analyzing executions to identify and resolve issues
Integrating Apps and Data
- Webhooks and form-based triggers
- Nodes for spreadsheets, email, and calendars
- Managing credentials and authentication for external services
- Step-by-step guide to the HTTP Request node
The AI Step
- Incorporating a model into a workflow
- Crafting prompts to ensure reliable output
- Generating structured output and automatically correcting malformed responses
- Classifying and extracting information from emails, forms, and documents
Decision Making and Actions
- Implementing branching logic with If and Switch nodes
- Creating approval steps to solicit human input before actions are executed
- Writing results to spreadsheets and dispatching notifications
Hands-on Lab: Initial End-to-End Workflow
- Managing the flow from an incoming request to a classified, recorded, and routed outcome
Day 2: Agents and Operations
Agents
- Defining an agent: model, memory, and tools
- Choosing between a fixed workflow and an agent
- Granting an agent access to company documents and spreadsheets
Action-Oriented Agents
- Distinguishing between tools that read data and those that write data
- Establishing system prompt rules: mandatory fields, prohibition of fabricated data, and prevention of duplicate records
- Testing an agent by attempting to disrupt its functionality
Communication Channels
- Setting up a chat interface for internal use
- Connecting the agent to a single messaging channel
Reliability
- Implementing error workflows, retries, and notifications
- Reviewing logs and execution history
- Managing model and execution costs
Data and Privacy
- Identifying what data leaves the company when a model is invoked
- Self-hosting and local models: instructor-led demonstration
- Creating a checklist for review with IT or legal teams
Constructing Workflows with AI
- Requesting an AI assistant to draft a workflow
- Reviewing the draft: analyzing the function of each node, data sources, and handling of empty states
Final Project and Next Steps
- Each participant finalizes and presents a workflow featuring an AI step and an agent applied to their own process
- Developing a maintenance plan and identifying future automation candidates
Deliverable: A functional workflow including an AI step, an agent connected to company data, error handling mechanisms, and a review checklist.
Requirements
- No prior programming experience is necessary
- Familiarity with using web applications and spreadsheets
- A clear understanding of one repetitive process within your own work
Target Audience
- Professionals in business, operations, and customer-facing roles
- Team leaders and process owners
- Non-developers seeking to build automations and AI agents using company data
Testimonials (3)
The trainer is patient and very helpful. He knows the topic well.
CLIFFORD TABARES - Universal Leaf Philippines, Inc.
Course - Agentic AI for Business Automation: Use Cases & Integration
Good mixvof knowledge and practice
Ion Mironescu - Facultatea S.A.I.A.P.M.
Course - Agentic AI for Enterprise Applications
The mix of theory and practice and of high level and low level perspectives
