What prior knowledge do I need?
This is an advanced workshop, the follow-up to Build your own AI agent (for beginners). You know the basics: connecting a trigger, model and action. If you don't yet, start with the beginners workshop first.
The next step after your first agent. In modules you build a complete, self-filling AI system: first a data pipeline that fills your knowledge base, then an agent that searches that data, calculates prices and emails quotes itself. Not a loose demo, but a working system that takes data in and does something with it. You work in the platform that fits your organisation: the open n8n or Microsoft Copilot Studio.
Not a loose exercise, but a complete system you built yourself: data in, and an agent that actually does something with it.
A data pipeline that processes multiple sources, enriches them with AI tags and writes them to a structured, searchable database.
An agent that searches your own knowledge base, calculates prices and sends a personalised quote as a clean email.
Both workflows as exports, plus your filled knowledge base, a summary and a list of sources to keep building.
Who is this workshop for?
You've built an agent before (trigger → model → action) and now want to build a complete system.
Departments that, after the first hands-on step, want to stand up a truly working AI system with their own data.
Who want to understand data ingestion, retrieval and tool use by building it themselves.
No basics yet? Start with Build your own AI agent (for beginners) →
So you can keep building in your own environment after the workshop
Your data-ingestion pipeline and your agent workflow, both as exports to keep working with.
The structured database you built yourself during Module 1.
The key points and framing per module, to read back at your own pace later.
A selection of tools, sources and next steps to keep building on your own.
The advanced concepts behind a working AI system, in plain language and applied right away in your platform.
The leap from one agent to an agentic system with its own data and real actions, and where that lands in practice.
How to fill a knowledge base automatically: multiple sources, access control and storage in a structured database.
How an agent searches your own data and answers from the source, instead of just making something up.
An agent with tools that calculates prices correctly and emails a quote, not one that only answers.
Step by step you build two workflows that connect: first the pipeline that fills your knowledge base, then the agent that acts on it. Guided, no one gets stuck.
Data comes in via a form and a spreadsheet connection. A password gate lets only authorised submissions through and you turn raw input into clean, reusable fields.
A language model generates tags automatically, then everything goes to a structured database: your self-filling knowledge base.
An agent with a role instruction and memory queries the knowledge base from Module 1 and works out prices correctly.
A mail tool sends the quote as a clean email to the client. From question box to something that actually does something.
Done with both? Module 3 (multi-agent, supervisor pattern) is ready to build on freely. Stefan walks around, no one gets stuck.
A clear division of roles: Anthonie sets the context per module, Stefan leads the build.
Background in Management, Economics & Law (Hanze), former COO at Ancora Health and an entrepreneur since 2010. Builds AI into existing organisations and sets the context per module: why you build this and where it lands.
Background in Human Technology (Hanze). Builds AI agents and AI infrastructure daily, and guides participants step by step as they build their own system.
This is an advanced workshop, the follow-up to Build your own AI agent (for beginners). You know the basics: connecting a trigger, model and action. If you don't yet, start with the beginners workshop first.
Both work. We align with you beforehand on which platform fits your organisation best: the open, flexible n8n or Microsoft Copilot Studio within your Microsoft 365 environment. The concepts and build steps are identical, only the tool differs.
Your own laptop. We provide the working environment (n8n or Microsoft Copilot Studio) and sandbox accounts. We align beforehand on any of your own credentials you'd like to use.
Advanced hands-on building needs good guidance, so we work with 8 to 20 participants so Stefan keeps everyone on pace.
The workshop costs from €149 per person (open enrolment) or from €2,500 excl. VAT for an in-company session. The exact price depends on group size and duration (2 to 4 hours, modular). Request a quote for a tailored proposal.
The builds use your chosen platform (n8n or Microsoft Copilot Studio), a language model, a mail connection and a database. For the session itself we provide sandbox accounts so you can build right away. If you want to use your own credentials, we align on that during the intake.
You take both workflows home as exports (templates), plus your filled knowledge base, a summary and a list of sources. So you can keep building in your own environment after the workshop.
Request a no-obligation quote for a tailored in-company session.