Our trainings are built for different audiences — from no-code basics to a production-grade developer track. Everything is available in Swedish, English and Spanish.
Free
12–15 minutes, no account and no payment. You leave with a ready-made instruction you can use right away.
A practical foundation course for understanding, designing and testing your first AI agent.
Go from single clever prompts to a scoped, testable and reusable way of working.
Pick the right problem, value the benefit and set up a safe pilot with clear ownership.
A shared language and a practical way to choose, design and control AI agents together.
Choose profitable and safe agent tasks, calculate net value and plan a small pilot.
Prioritise agent areas, set simple governance and run a measurable pilot without unnecessary tech or cost.
For anyone new to MCP: what it is, when it makes sense and which questions to ask first.
A follow-on to the foundation course: build, test and cost-check a simple meeting assistant.
Build a quality reviewer in ChatGPT or Claude Projects, measure quality and calculate the value.
Prioritise the portfolio, build a pilot, run acceptance tests and decide to stop, adjust or scale.
A shared build where roles, sources, permissions and testing fit together.
Immediately useful assistants for briefs, quotes, delivery and follow-up.
Build a shared assistant in ChatGPT or Claude — with sources, ownership and a value calculation.
Turn meeting notes into clear decisions, actions, owners and open questions — without invention.
Review tech and content, build an SEO assistant and create a 30-day plan for AI visibility.
Build a source-aware search agent that finds current sources, weighs evidence and cites correctly.
Design, build and test a controlled agent system with tools, permissions and human approval.
Build a practical AI agent in Make or n8n – without traditional programming.
We build the courses one at a time. Sign up to be notified as soon as a course opens. The prices below are preliminary.
Build a research assistant that separates source, summary, assumption and uncertainty.
Preliminary price — not final and not an active offer.
Judge whether an agent actually does the right thing — with test cases, metrics and human review.
Preliminary price — not final and not an active offer.
Let the agent answer from your own documents — with structure, permissions and clear sources.
Preliminary price — not final and not an active offer.
Set permissions, approval steps and traceability before the agent acts for real.
Preliminary price — not final and not an active offer.
Connect your assistant to everyday tools — step by step and without programming.
Preliminary price — not final and not an active offer.
Faster replies with the right tone, clear limits and a safe handover to a human.
Preliminary price — not final and not an active offer.
Produce content in your own voice — with fact-checking, structure and a final human review.
Preliminary price — not final and not an active offer.
The field moves fast. Every three months we update all course content and clearly show what is new, changed or removed — at no extra cost.