Course Outline
Day 1: Workflows and the AI step
Introduction
- When to automate, when to use an AI step and when to use an agent
- n8n in context: cloud and self-hosted, and what each means for company data
n8n essentials
- Workflows, nodes, triggers and executions
- Moving data between nodes by dragging fields, without typing syntax
- Reading an execution to find and fix a problem
Connecting apps and data
- Webhooks and form triggers
- Spreadsheets, email and calendar nodes
- Credentials and authentication for external services
- The HTTP Request node, step by step
The AI step
- Adding a model to a workflow
- Writing prompts for reliable output
- Structured output and automatic correction of malformed answers
- Classifying and extracting information from emails, forms and documents
Decisions and actions
- Branching with If and Switch
- Approval steps: asking a person before an action is taken
- Writing results to a spreadsheet and sending notifications
Hands-on lab: first end-to-end workflow
- From an incoming request to a classified, recorded and routed result
Day 2: Agents and operations
Agents
- What an agent is: model, memory and tools
- Fixed workflow or agent: how to choose
- Giving an agent access to company documents and spreadsheets
Agents that act
- Tools that read and tools that write
- System prompt rules: required fields, no invented data, no duplicate records
- Testing an agent by trying to break it
Channels
- Chat interface for internal use
- Connecting the agent to one messaging channel
Reliability
- Error workflows, retries and notifications
- Logs and execution history
- Controlling model and execution costs
Data and privacy
- What leaves the company when a model is called
- Self-hosting and local models: instructor demonstration
- A checklist to review with the IT or legal team
Building workflows with AI
- Asking an AI assistant to draft a workflow
- Reviewing it: what each node does, where the data comes from and what happens when it is empty
Final project and next steps
- Each participant completes and presents a workflow with an AI step and an agent on their own process
- Maintenance plan and next candidates for automation
Deliverable: a working workflow with an AI step, an agent connected to company data, error handling and a review checklist.
Requirements
- No programming experience required
- Comfort using web applications and spreadsheets
- A clear picture of one repetitive process in their own work
Audience
- Business, operations and customer-facing professionals
- Team leads and process owners
- Non-developers who want to build automations and AI agents on company data
Testimonials (3)
Strong confirmation that current LLMs are deployable for customer projects when the proper setup is in place.
Oliver Fromm
Course - Agentic AI for Enterprise Applications
Machine Translated
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
The mix of theory and practice and of high level and low level perspectives
