Futurepedia

Free Course: Introduction to AI Agents

Learn how to build powerful AI agents using no-code tools, integrating platforms like OpenAI, Google Calendar, and APIs to create dynamic, automated workflows.

Unleash the full potential of your business with the collection of HubSpot's AI tools, seamlessly integrated throughout the entire customer platform.

Get a Demo

Your first AI agent starts here...

This hands-on crash course takes you through the fundamentals and practical application of building AI agents using the n8n platform. You'll start by understanding what AI agents are, how they work, and what makes them distinct from traditional automation. With a focus on accessibility, the course shows how to build a fully functioning AI agent from scratch—no coding experience required. You'll learn how to configure an agent's brain with OpenAI, set up memory for contextual awareness, and connect a variety of tools like Google Calendar, Gmail, and even custom APIs using HTTP requests. By the end, you'll have the skills to build your own personalized AI assistant that adapts and acts on your behalf.

What You Will Learn
- The core concepts behind AI agents and how they differ from automations
- Key components of an AI agent: brain (LLM), memory, and tools
- How to use n8n’s visual interface to build agents with no coding required
- Connecting and using APIs and HTTP requests in your workflows
- Setting up tools like Google Calendar, Gmail, and custom APIs
- Writing structured prompts to guide agent behavior
- Testing, troubleshooting, and refining AI workflows
Who This Course is For
- Beginners curious about AI agents with no prior coding experience
- Professionals seeking to automate personal or business tasks
- Makers and tech enthusiasts interested in building custom AI workflows
- Anyone looking to explore the capabilities of n8n and AI integration
Requirements
- A free or trial account on n8n.io
- Basic familiarity with tools like Google Calendar and Gmail is helpful but not required
- Optional: OpenAI account and API key for LLM integration

Section 1

Welcome to the AI Agents Course

Section 2

Introduction to AI Agents

2.1

Introduction to AI Agents

What actually makes an AI agent different from a regular automation? Get the full breakdown of how agents think, remember, and act using the brain-memory-tools framework — plus why even complex multi-agent systems are just repeating the same simple concepts.

2.2

Preparing to Build Our First AI Agent

APIs and HTTP requests sound intimidating, but they're surprisingly simple once you see how they actually work. Walk through both concepts in plain English, then see exactly what kinds of agents — from email assistants to travel planners — you can build with just these fundamentals.

Section 3

Building AI Agents with n8n

3.1

Getting Started with n8n

n8n lets you build full AI agents by dragging and dropping — no code required. Get set up from scratch, learn how nodes and connections work, and see why n8n's dedicated AI agent node gives you plug-and-play access to the brain, memory, and tools your agent needs.

3.2

Building An Agent with n8n

Connect your agent's brain (OpenAI), give it memory so it can hold a conversation, and start wiring up real tools — Google Calendar, weather APIs, Gmail, Google Sheets, and even a custom air quality check using an HTTP request. Plus, learn how to write the prompt that tells your agent exactly what to do.

3.3

Testing Our n8n Agent

Time to hit "test" and see what happens — errors included. Walk through the fastest way to debug any issue (hint: screenshot it and drop it into ChatGPT), then watch the finished agent deliver a personalized trail recommendation straight to your inbox.

Section 4

Building a Personal Assistant Agent with n8n

4.1

Build a Daily Personal Assistant in 1 Workflow

A completely different n8n build from scratch — this time a personal assistant that reads your client emails, checks your calendar, and proposes meeting times without you lifting a finger. Every step is walked through slowly so you can follow along even if the previous section moved too fast.

4.2

Give Your Agent an AI Model

Your agent needs a brain, and here's how to pick one. Connect OpenAI to your n8n agent, grab your API key step by step, and understand the tradeoff between powerful models and cost-effective ones so you're not overspending on every request.

4.3

Give Your Agent Memory and Tools

Add memory so your agent can track conversations, then connect it to Google Calendar and Gmail with real credentials. Learn how to filter emails by label, set time-based expressions, and rename your tools so everything stays organized as your agent gets more complex.

4.4

Test and Publish Your Agent

Stack multiple triggers, run your first test, troubleshoot errors in real time, and watch your agent analyze client emails, cross-reference your calendar, and deliver a formatted briefing with a proposed meeting time — all on autopilot.

Section 5

Make AI Agents

5.1

Introduction to Make AI Agents

Make is a visual automation platform with over 2,000 app connections — and now it has full AI agent capabilities. See exactly what we're building: an agent that automatically repurposes YouTube scripts into LinkedIn posts and YouTube Shorts without any manual copy-pasting.

5.2

Make AI Agent - The Basics

Set up your first Make AI agent from scratch — pick your AI model, connect your OpenAI API key, and write the system prompt that tells your agent exactly how to repurpose content. Everything is walked through step by step so you can follow along on a free account.

5.3

Make AI Agent - The Context

Modules handle single actions; scenarios handle entire multi-step workflows. Build your first scenario — one that grabs the full text from a Google Doc using its file ID — and connect it to your agent so it can pull scripts automatically.

5.4

Make AI Agent - Tools & Senarios - Part 1

Modules handle single actions; scenarios handle entire multi-step workflows. Build your first scenario — one that grabs the full text from a Google Doc using its file ID — and connect it to your agent so it can pull scripts automatically.

5.5

Make AI Agent - Tools & Senarios - Part 2

Wire up the rest of your agent's toolkit: a Google Docs module to create new documents, a LinkedIn integration to auto-post, and a Slack notification so you know the moment a new YouTube Short script is ready to record.

5.6

Make AI Agent - Finalize, Test and Publish

Connect everything to a Google Drive trigger, run the full workflow end to end, and watch your agent repurpose a script, post to LinkedIn, create a new doc, and ping you on Slack — all from a single file drop. Then set it to run daily and let it work in the background.

Section 6

A Tour of Other Top AI Agent Platforms and Features

6.1

Build a AI Support Agent Step-by-Step

Build a business-ready customer support agent in about 12 minutes using Chatbase — no coding, no technical skills. Crawl any website to build a knowledge base, customize the branding, pick your AI model, and deploy it as a chat widget or full help page experience.

6.2

Building a Productivity AI Agent in Notion

Notion's new AI agents aren't chatbots — they're AI teammates that can search your entire workspace, update dozens of pages, and deliver a summary document with links to everything they changed. See four real demos: project kickstarts, SOP audits, knowledge base updates, and a daily refresh workflow that saves nearly an hour every morning.

Conclusion

You're Ready to Build

You just went from "what even is an AI agent?" to building real, working agents across three different platforms — without writing a single line of code.

The brain-memory-tools framework that powers every AI agent, from simple weather bots to complex multi-agent systems
The difference between automations and agents — fixed steps vs. dynamic reasoning that adapts on the fly
APIs and HTTP requests — the two concepts that scare most people off are just structured ways for software to talk to each other
How to write structured prompts that tell your agent exactly who it is, what it does, and how to behave

n8n Trail Assistant: Checks your calendar, pulls weather and air quality data, references a personal spreadsheet, and delivers a trail recommendation to your inbox
n8n Personal Assistant: Reads client emails, cross-references your calendar, and proposes meeting times — all on autopilot
Make Repurposing Agent: Turns YouTube scripts into LinkedIn posts and Short-form video scripts, auto-posts them, and pings you on Slack when they're ready
Chatbase Support Agent: A customer support bot trained on an entire website's knowledge base, deployed in about 12 minutes
Notion AI Agent: Audits your workspace, updates outdated references across dozens of pages, and delivers summary reports — saving nearly an hour every morning

No matter the platform or use case, every agent comes back to the same five steps:

Pick your brain (the LLM)
Give it memory (so it has context)
Connect your tools (so it can act)
Write a clear prompt (so it knows what to do)
Test until it works the way you want

Once you've internalized that pattern, you can build agents for research, customer support, sales workflows, content production, scheduling, data analysis — and beyond.

Start Simple. Ship Something Real.

Pick one task you do repeatedly — something that eats 30 minutes of your day — and automate it with a single agent. Get that working reliably. Then layer in complexity: more tools, better prompts, multi-step scenarios, maybe even multi-agent systems. But always start with the simplest thing that works.

The tools are accessible, the platforms have free tiers, and you now have the skills to build. The only thing left is to pick your first workflow and go.

Ready to Get Started?

Breeze Assistant is your AI go-to-market expert, built into HubSpot. Start using these prompts today.

Get started free