Description
n8n has become one of the fastest routes from "we should try an agent" to something running in production. This day covers the AI Agent node properly: giving an agent tools, memory and retrieval over your own documents, then wiring it into the systems you already run.
What you will leave with
- Build an agent that selects and calls tools to complete a task
- Add memory so an agent holds context across a conversation
- Ground an agent on your own documents with a vector store
- Chain agents and sub-workflows for multi-step work
- Put limits and approvals around what an agent may do
Agenda
| Agent vs workflow: when the model should decide the path | |
| The AI Agent node — models, prompts, and the tool loop | |
| Giving agents tools: HTTP, database and custom sub-workflows | |
| Memory: what to keep, what to summarise, what to drop | |
| Retrieval — embeddings and vector stores over your own content | |
| Multi-step and multi-agent patterns without losing control | |
| Guardrails, approvals and cost limits | |
| Capstone: an agent handling a real request end to end |
Who it is for
Anyone who has built a basic n8n workflow and wants to add genuine reasoning and tool use to it.
Prerequisites
Basic n8n familiarity (our n8n Fundamentals half day covers it) and an API key for a model provider.
Format
- Duration: Full day (09:00–17:00)
- Level: Intermediate
- Group size: maximum 10 attendees
- Delivery: live online, or in person by arrangement
Need this for a whole team, or extended to full length on your own systems? We deliver every workshop as a bespoke on-site course too — talk to us.
Reviews
There are no reviews yet.