⊛ Intermediate

n8n AI Agents: Build Intelligent Automation

Build AI agents in n8n that reason, remember, and act. Connect LLMs, add persistent memory, use built-in tools, implement RAG with vector stores, and orchestrate multi-agent pipelines.

📚 14 lessons 🕐 195 min total ✓ Free

The AI Agent Node: How n8n Thinks

Understand how n8n's AI Agent node works: the Tools Agent architecture, sub-node model, and the reasoning loop that drives every agent decision.

Connecting an LLM: OpenAI, Claude, Gemini, and Ollama

Learn how to connect language models to your AI Agent — from OpenAI and Claude to local Ollama — and what to consider when choosing between them for cost, speed, and privacy.

Your First AI Agent: A Research Assistant

Build your first AI Agent end-to-end: a research assistant that searches the web, reasons through results, and answers questions — using the built-in n8n chat interface.

Memory: Giving Your Agent Context

Learn how to add memory to an AI Agent — from volatile Simple Memory to persistent Postgres and Redis — so your agent can hold multi-turn conversations across sessions.

Tools: Extending What Your Agent Can Do

Expand your AI Agent's capabilities: connect built-in utilities, app integrations, and web search tools, and learn to write tool descriptions that guide the agent to the right choice.

Lightweight AI Nodes: LLM Chain, Extractor, Classifier, Sentiment

Learn n8n's four lightweight AI nodes — Basic LLM Chain, Information Extractor, Text Classifier, and Sentiment Analysis — and know when to use them instead of the full AI Agent.

Building a Customer Support Chatbot

Build a multi-turn customer support chatbot: Chat Trigger, AI Agent with persistent memory, a knowledge base tool, and a human escalation path for questions the agent cannot resolve.

Document Ingestion: Loading Data for RAG

Build the ingestion half of a RAG system: load documents, chunk them with a text splitter, generate embeddings, and store the vectors in a database ready for semantic retrieval.

Vector Stores and Retrieval: Making Your Agent Smart

Understand vector embeddings and semantic search, compare the vector stores supported in n8n, and connect your knowledge base to an AI Agent using the Vector Store Tool.

Building a RAG Knowledge Base Chatbot

Build a complete end-to-end RAG system: an ingestion pipeline to load your documents into a vector store, and a chat workflow where an AI Agent retrieves and answers questions.

AI Email Triage and Auto-Reply

Build an AI email triage workflow: classify incoming emails with a Text Classifier, auto-draft support replies with an AI Agent, and route sales leads to your CRM automatically.

Agentic Content Creation Pipeline

Build an agentic content pipeline that takes a topic brief, researches it with web search, drafts a full article, extracts SEO metadata, and routes the draft for human review.

Multi-Agent Patterns: Sub-Agents and Delegation

Learn how to design multi-agent workflows: an orchestrator that delegates to specialist sub-agents using the Workflow Tool, with error handling and execution monitoring for production use.

MCP: n8n as Server and Client

Learn how n8n's MCP integration works: expose workflows as MCP tools using the MCP Server Trigger, and give AI Agents access to external MCP servers using the MCP Client Tool.

Start Lesson 1 →