AI Agents at Work: No-Code Automations with n8n, RAG & MCP

About this course

This course contains the use of artificial intelligence.

By the end of this course, you will have working AI agents doing real jobs in your workday — triaging your inbox, answering from your documents, chasing your meetings — each one gated, tested and under your control, and you will have built every one of them without writing code.

You already know how to prompt a chatbot. This course is the step after: delegating whole jobs to AI agents that run in n8n while you stay in charge of anything that matters.

WHAT YOU WILL BUILD1. An Inbox Triage Agent that classifies your email, drafts replies in your voice — and is physically unable to send anything until you approve it.2. A Document Q&A Agent (RAG) that answers questions from your own documents, cites the exact source for every answer, and says "not covered" instead of inventing one.3. A Meeting Follow-Up Agent that turns transcripts into decisions, owners and dates — flagging what was never said instead of making it up — and chases open actions with drafts you approve.4. A Research & Lead Agent that produces sourced briefs from the live web and scores incoming leads with its reasons written out.5. A capstone agent on your own real job — designed on one page, tested against a written test set, timed against your old way of doing it, and ready to present to your manager.

Plus the MCP lab: Your agents gain tools through the Model Context Protocol, and you expose one of your own workflows as a tool other AI clients can call — with authentication on and its approval gate intact.

THE SAFETY SPINE — WHY THIS COURSE EXISTSMost agent courses teach you to build agents that can. This one teaches you to build agents you can trust at work. Every build in this course is finished with the Safety Spine:

Read the full description on Udemy on Udemy (opens in a new tab)

Prices, discounts and availability are set by Udemy. We may earn a commission when you purchase through links on this site.