Workshops: AI Automation & Agents

n8n Workshop: AI Automation for Businesses

Over two days your team learns to automate business processes with n8n and to integrate AI models such as ChatGPT, Claude or Mistral where they add value. You build near-production workflows with error handling and learn how to run n8n in a privacy-friendly way.

Who is this workshop for?

  • Automation and process teams
  • IT departments looking for an alternative to cloud-only tools
  • Tech-savvy staff from operations, marketing or sales
  • Teams already using Make or Zapier who want more control
  • Organisations with strict data residency requirements

What you will take away

  • You build workflows with triggers, branches, loops and data transformation in n8n.
  • You integrate language models via AI nodes and produce structured, reliable output.
  • You build an AI agent in n8n that uses tools such as databases and HTTP calls.
  • You connect systems without a ready-made node using HTTP requests and webhooks.
  • You set up error handling, credential management and monitoring.
  • You know the cloud and self-hosting options and their security implications.

Agenda

  1. Module 1

    Day 1: n8n fundamentals

    • Interface, nodes, executions and data structure
    • Triggers: schedule, webhook, email, form
    • Transforming data with expressions and code nodes
  2. Module 2

    Day 1: AI in the workflow

    • Connecting models from OpenAI, Anthropic and Mistral
    • Structured output with JSON schema
    • Exercise: classify and route incoming requests
  3. Module 3

    Day 2: AI agents in n8n

    • AI agent node with tools and memory
    • Knowledge retrieval via a vector database
    • Setting limits: tool permissions and approval steps
  4. Module 4

    Day 2: Integrations and APIs

    • HTTP request, authentication, pagination
    • CRM, ticketing, Microsoft 365 and Google Workspace
    • Exercise: your own end-to-end process
  5. Module 5

    Day 2: Operations

    • Error workflows, retries, alerts
    • Self-hosting with Docker, updates, backups
    • Credentials, roles and documentation

Why n8n?

n8n combines the ease of a visual workflow editor with the flexibility of code. Where a ready-made node is missing, an HTTP request or a JavaScript node fills the gap. For many companies the deciding factor is that n8n can run on their own servers or in an EU cloud, so data does not pass through third-party automation platforms.

Note that n8n's licence model differs from classic open-source software. We explain what that means for internal use.

A practical example: incoming contact form requests are classified by topic and urgency by a language model, enriched with CRM data and routed to the right person – including a draft reply. Participants rebuild exactly this kind of flow with their own systems on day two.

AI models we integrate

n8n is model-agnostic. We show the same workflow with different models so your team can compare quality, speed and cost.

  • OpenAI models (as used in ChatGPT)
  • Claude by Anthropic
  • Mistral as a European provider
  • Google Gemini
  • Local models, e.g. via Ollama, for highly sensitive data

Data protection in automated processes

Automations often process personal data at scale. We cover data minimisation, logging, retention and which data may be sent to external AI APIs. For workflows touching employee data, we recommend involving the works council early.

Frequently asked questions

Do participants need coding skills?

A basic understanding of data and APIs helps but is not required. We use code nodes selectively and explain every step.

Cloud or self-hosting – what do you recommend?

n8n Cloud is the fastest way to start. For sensitive data or many workflows, self-hosting often makes more sense. We help you decide and, on request, with the setup.

Can we bring our own processes?

Yes, we strongly recommend it. On day two, groups build a real process from your organisation.

Workshop

Get in touch

Which process would you automate first with n8n and AI?

Tell us a little about your team – we will get back to you within 24 hours with a proposal for content, format and dates.

Prefer to talk?

We only use your details to handle this request. See our privacy policy.