Course Outline
Day 1: Workflows and the AI step
Introduction
- Determining when to automate, when to incorporate an AI step, and when to deploy an agent
- n8n in context: cloud vs. self-hosted options and their implications for corporate data
n8n essentials
- Understanding workflows, nodes, triggers, and executions
- Transferring data between nodes via drag-and-drop, eliminating the need for syntax typing
- Analyzing execution logs to identify and resolve issues
Connecting apps and data
- Implementing Webhooks and form triggers
- Integrating Spreadsheets, email, and calendar nodes
- 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
- Achieving structured output and automatically correcting malformed responses
- Classifying and extracting information from emails, forms, and documents
Decisions and actions
- Implementing branching logic with If and Switch nodes
- Establishing approval steps: requiring human verification before actions are executed
- Recording results in spreadsheets and dispatching notifications
Hands-on lab: first end-to-end workflow
- Processing an incoming request to a classified, recorded, and routed outcome
Day 2: Agents and operations
Agents
- Defining an agent: the role of model, memory, and tools
- Choosing between a fixed workflow and an agent
- Enabling agent access to corporate documents and spreadsheets
Agents that act
- Differentiating between tools that read and tools that write
- System prompt rules: enforcing required fields, prohibiting invented data, and preventing duplicate records
- Stress-testing an agent by attempting to break it
Channels
- Setting up a chat interface for internal use
- Connecting the agent to a single messaging channel
Reliability
- Configuring error workflows, retries, and notifications
- Utilizing logs and execution history
- Managing model and execution costs
Data and privacy
- Assessing data egress when a model is invoked
- Demonstration of self-hosting and local models by the instructor
- Providing a checklist for review with IT or legal teams
Building workflows with AI
- Prompting an AI assistant to draft a workflow
- Reviewing the draft: analyzing node functions, data sources, and empty-state handling
Final project and next steps
- Participants finalize and present 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 linked to corporate data, error handling mechanisms, and a review checklist.
Requirements
- No programming background required
- Familiarity with using web applications and spreadsheets
- A clear understanding of one repetitive process within their own workflow
Audience
- Business, operations, and customer-facing professionals
- Team leads and process owners
- Non-technical users looking to build automations and AI agents using corporate 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
