Employees with some AI experienceIntermediate140–160 min

Build a reusable AI assistant in ChatGPT and Claude

Build a quality reviewer through eight modules with Tallvik’s fictional case, five runnable tests with model answers, a value calculation and a 30-day pilot package. Your own answers, test responses and calculation assumptions are included; the worked example is not.

After the course you can:

  • ✓Move from one-off prompt to reusable workflow
  • ✓Design a reviewer with a fixed output format
  • ✓Set your own evaluation criteria

USD 49 incl. applicable taxes

Your final price and currency are shown at checkout based on your billing country.

12 months of access · updates included

Buy the course – USD 49Try it free first – 18 min, no account
  • Payment via Stripe
  • 12 months of access
  • Certificate with verification link
  • Swedish company, reg. no. 559599-6850

For employees with some experience using AI · 140–160 minutes

What you will build

AI assistant

You will learn to: Build an AI assistant in ChatGPT or Claude with your own material and tests

About the course

This course is for people who already use ChatGPT, Claude or similar tools but have not yet built a structured workplace assistant. You compare a one-off prompt, saved template, project, agent and automated workflow, and learn when additional structure creates real value.

In the practical exercise, you build a quality reviewer in ChatGPT Projects or Claude Projects. You define review criteria, add approved examples, create a fixed response format and run five test cases. The course concludes with a value calculation and a monthly maintenance plan.

This course is for you if you

  • ✓use AI at work but want to reuse effective solutions
  • ✓need clearer quality assurance and traceability
  • ✓want to compare ChatGPT and Claude using your own task

After the course you can

  • ✓decide when a project creates more value than a standard prompt
  • ✓build a workplace assistant in ChatGPT Projects or Claude Projects
  • ✓create a simple test suite
  • ✓measure time savings and quality
  • ✓maintain the solution without unnecessary costs

You leave with

  • ✓an assistant card
  • ✓copy-ready project instructions
  • ✓five test cases
  • ✓a value calculation
  • ✓a 30-day pilot plan

What you will be able to do

  • ✓Move from one-off prompt to reusable workflow
  • ✓Design a reviewer with a fixed output format
  • ✓Set your own evaluation criteria
  • ✓Compare ChatGPT and Claude objectively
  • ✓Calculate time saved, fees and maintenance
  • ✓A personal course certificate once you have completed all parts and passed the knowledge test – with its own verification link, easy to share on LinkedIn or add to your profile.

Course content

  1. 1When a smart prompt is no longer enough
  2. 2Popular assistants for everyday work
  3. 3Score the task before you build
  4. 4Build the instructions for the quality reviewer
  5. 5Put instructions, sources and examples in the right place
  6. 6Test the same five tasks in both tools
  7. 7Calculate time saved, gross value, and net value
  8. 8Stress-test, maintain, and compile your pilot summary
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.

More about us →

Source-checked content

The course is reviewed against official documentation and established standards. The organizations below are examples of sources whose material the content builds on.

  • OpenAI
  • Anthropic
  • NIST
  • OWASP GenAI Security Project
  • Swedish Authority for Privacy Protection (IMY)
Last checked:
September 19, 2026
Next scheduled review:
December 19, 2026
How we review the content