For beginners – and anyone who has started vibe coding

Build with AI – step by step, at your own pace

You don't need to know how to code. But building with AI isn't as easy as it looks. It's easy to get stuck, lose track, or end up doing the same thing over and over. Here you'll learn to do it right from the start: with a clear idea, small steps and tests that show it works. That saves you both time and money.

Try free – 18 minutesSee the courses

Made for beginners · No coding required · Courses in English

  • Payment via Stripe
  • 12 months of access
  • Certificate with verification link
  • Swedish company, reg. no. 559599-6850
  • Courses reviewed every three months
This is what the course looks like – short parts, your own exercises, instant feedback

Good to know before you start

What you need

  • A task you want to solve with AI. Feel free to start with something small.
  • An account with an AI service, such as ChatGPT or Claude.
  • A few hours per course, plus some time to try things on your own.
  • A willingness to test and check. AI makes mistakes – in the courses you learn to prevent them and find them.

What you get

  • Common pitfalls, and how to avoid them before they cost you time or money.
  • Templates, checklists and exercises you can use right away.
  • A course certificate when you finish.

Where are you today?

I haven't started yet

You've heard about AI and maybe tried ChatGPT, but you're not quite sure where to begin. You're in the right place – this is where most people start.

Start with the free course (about 18 minutes)

I've started vibe coding

You've built something with the help of AI and want to take the next step. Learn how to make what you've built work in the long run, and how to let AI assistants and AI agents do work for you.

See the next step

Already know what you want to learn? See all courses

How to move forward – step by step

  1. 1

    Step 1 – Create with AI (vibe coding)

    You describe what you want to create, and AI helps you build it – for example a website, an app, or a simple tool.

    Courses: Vibe Coding with ChatGPT and ClaudeVibe Coding in Practice

  2. 2

    Step 2 – Let AI work for you (AI assistants)

    An AI assistant handles recurring tasks for you, following the instructions and rules you set.

    Courses: AI Agents for BeginnersBuild Your First AI Assistant

  3. 3

    Step 3 – Automate (AI agents)

    An AI agent can carry out several steps on its own and use the tools you give it access to.

    Course: Build AI Agents Without Code

With vibe coding, you create with the help of AI. With AI assistants and AI agents, you let AI do work for you.

What a course looks like

Short parts – you always know where you are
Instant feedback – you see why
Course certificate – share on LinkedIn
Michael Hammarstedt, founder of AI-agentutbildning
The courses are created by Michael Hammarstedt, founder of AI-agentutbildning. I'm not a programmer by background – I started with vibe coding myself. That's why the courses explain everything step by step, with no prior knowledge needed.

Michael Hammarstedt

Founder, AI-agentutbildning

More about us →

Why you can feel safe

A real company

Iqlear Dev, Byggkomposit Sverige AB, reg. no 559599-6850, Ulrika, Sweden

Secure payment

Payments are handled by Stripe. We never see your card details.

Try it first

The free course takes 18 minutes and requires neither an account nor a card.

Reply within two working days

kontakt@aiagentutbildning.se

Always up to date

We review every course at least every three months and update it when tools or rules change. Updates are included for your entire access period.

Try a real course for free

Not sure whether the courses are right for you? Start for free and experience how a course works.

You work on a realistic task, make your own choices, and get immediate feedback. At the same time, you build something useful that you can keep using after the course.

It is a complete mini-course, not just a preview.

  • Seven short sections in about 18 minutes
  • Interactive exercises with immediate feedback
  • A personal result based on your answers
  • A complete AI instruction to copy
  • Works on desktop, tablet and mobile
  • No account or payment required
Start the free course

AI assistant or AI agent – what’s the difference?

You do not need to know all the terminology before you begin. Here you get a simple explanation and see how solutions can develop from a clearly instructed assistant into a more independent AI agent.

AI assistant

You start the work, provide the task and review the result. It is well suited to tasks such as text review, research, meeting summaries and report drafts.

AI agent

Works toward a goal and can choose between permitted steps or tools. It requires clear permissions, review checkpoints and stopping rules.

From a one-off prompt to an automated workflow

  1. 1One-off prompt You ask for a summary of a text once.
  2. 2Saved template You save the same instruction and reuse it every week.
  3. 3Reusable AI assistant A project with an instruction and approved material that a person starts and reviews.
  4. 4AI agent with tools and several steps A goal is broken into steps where the agent chooses among permitted tools and stops at the checkpoints.
  5. 5Automated workflow A new form response starts a defined flow that runs the same way every time.

More autonomy does not remove the need for human oversight. The more steps a solution may choose on its own, the clearer its permissions, review checkpoints and stopping rules must be.

Not every task needs an agent. A well-scoped AI assistant is often the simplest and most cost-effective solution.

Read the guide to AI assistants and AI agents →

Popular topics right now

A selection of courses in areas many people want to get started with right now.

View all courses

How you learn in our courses

Our courses are built on four principles that research shows work. The goal is for you to use what you learn in your work right away – not just recognize it.

  • You do, not just read

    In every course you work through exercises based on your own situation instead of only reading.

    The research: In a review of 225 studies, active learning led to better exam results, and students in traditional lectures were 1.5 times more likely to fail. (Freeman et al., PNAS 2014)

  • You test yourself and get instant feedback

    Check questions with explanations and a knowledge test at the end.

    The research: Retrieving knowledge from memory leads to better retention than rereading – especially with feedback. (Rowland, Psychological Bulletin 2014; Dunlosky et al. 2013)

  • Short parts in a clear order

    The material is split into short parts you take at your own pace, and you always see where you are.

    The research: Breaking new material into smaller segments is a well-established principle for learning on screen. (Mayer, Cambridge Handbook of Multimedia Learning)

  • Concrete examples to learn from

    Realistic example businesses, ready-made instructions, and model answers to compare with.

    The research: Beginners learn more from studying worked examples than from solving everything on their own from the start (the worked-example effect, Sweller).

Sources: Freeman et al. (PNAS 2014), Rowland (Psychological Bulletin 2014), Dunlosky et al. (2013), Mayer (Cambridge Handbook of Multimedia Learning). → Read more about how we build and review our courses

What you take away

Depending on the course, you can finish with:

  • a complete description of what the AI assistant should do, which material it may use and where its limits lie
  • a copy-ready instruction for ChatGPT or Claude
  • a set of realistic test cases
  • clear quality criteria and checkpoints
  • a simple value calculation with estimated time saved, cost, and net value
  • a plan for a pilot, implementation and follow-up
  • ready-made recipes for AI assistants, AI agents, and automated workflows that you can adapt to your organization
  • a personal course certificate with its own verification link once you have completed the course – easy to share on LinkedIn or add to your profile

Your work is saved throughout the course and can be copied, downloaded or printed.

Find the right agent idea

The AI library collects concrete ideas for real work, with required material, checkpoints and risk level. Three recipes are free to use right away.

Open the AI library

Courses for different roles and needs

The courses are designed for people who want to use AI in practical ways at work without needing to be programmers. There are courses for:

  • employees who want to build reusable work assistants
  • managers who need to assess value, risk and accountability before starting a pilot
  • teams that want shared workflows, instructions and test criteria
  • solo entrepreneurs and small businesses that want to save time on recurring tasks
  • people ready to continue with automation, Make, n8n, MCP and more advanced AI agents

You do not need programming skills. The courses show you how to start with existing AI tools and expand only when the task, quality requirements and value are clear.

Source-checked and regularly reviewed content

AI tools, features and prices change quickly. That’s why we review the content regularly and label information that may change with the date it was last checked.

When a course describes external features, security principles or regulation, we primarily use official sources from organizations such as OpenAI, Anthropic, NIST, OWASP and the European Union. The content is reviewed at least once every quarter.

These sources do not replace your organization’s own policies, legal assessment or security requirements. They show what supports the factual content of the course and when it was last checked.

Sources

  • OpenAI
  • Anthropic
  • NIST
  • OWASP
  • European Union
Last checked:
September 19, 2026
Next scheduled review:
December 19, 2026
How we review the content

Practical courses in AI agents and AI assistants

From one-off prompts to reusable work assistants

It is easy to ask an AI tool to write a text or summarize a document. The greater value comes when you can create a reusable AI assistant that follows clear instructions, works from approved material and returns results in a defined format.

Our courses teach you to distinguish between tasks that suit AI, tasks that need a narrower scope and tasks that must always involve human judgment. You build the instruction step by step, decide which sources the assistant may use and define checkpoints for quality, uncertainty and sensitive information.

You also test the assistant against realistic situations, such as incomplete material, conflicting information, sensitive data or a response that does not follow the required format.

The goal is not to hand responsibility over to AI. The goal is to create a useful way of working in which AI handles repeatable groundwork and a person reviews, evaluates and approves the result.

See all courses and filter by what you want to build

In the catalog, you can filter by solution type, audience, level, and status, and sort by recommended order, level, or price.

Explore all courses