Advanced · Critical thinking and your own skills · module 130 of 155
Final project: your personal AI practice portfolio
Bring together your method cards, test sets, data agreements and reflection in one portfolio that you can show and keep using.
After this lesson
You put together a practice portfolio with three worked-out applications, their checks and limits, and a plan to keep your skills up to date.
In view · How it connects
Your AI practice portfolio
Purpose, task and result, with fictional or cleared examples.
Method cards, test sets, data agreements and the chosen tool.
What worked, what you will do differently and when you will review the portfolio.
Part 1 of 3 · Three applications
Your preference for less movement is on. Choose each part yourself.
Choose three applications from your own practice
A practice portfolio shows what you can do with AI, and how carefully you work. Choose three applications that you worked out in this course or use now, preferably from different areas: for example a document workflow (lesson 110), an automation (lesson 115) and a media production (lesson 125). Use fictional or cleared examples; a portfolio you show must not contain confidential data. For each application, describe the goal, the task and the result.
Show your method, not just the result
For each application, add your method card: source, prompt, checks, human decision points and second route (lesson 100). Add the test set with the results, and note which errors your checks caught (lesson 113). Describe which data you did and did not use and which agreements applied (lesson 118). Also record which model or tool you used and why (lesson 128). That way someone else can assess, take over and repeat your work.
Reflect and plan your next step
For each application, write briefly what worked well, what surprised you and what you would do differently. Where did you trust AI too quickly, and which check helped (lesson 126)? Choose one skill you want to develop further and plan a concrete trial with a date (lesson 95). Review your portfolio every six months: tools and agreements change, and your methods must keep up. This portfolio is self-study and not a recognised diploma, but it shows concretely how you work with AI.
New practice example
From explanation to a usable result.
A practice portfolio shows not only results, but also your working method, checks, agreements and reflection.
Fictional practice situationFictional: an employee wants to put together a portfolio of three AI applications: action lists from meetings, a monthly overview of invoices and a flyer for an open day. So far, he has only kept the end results.
View a possible solution
This is how it can be done
For each application he adds: the method card with prompt and checks, a small test set with outcomes, the data agreements and a short reflection. He replaces the real invoices with fictional ones. A colleague tries to repeat the action lists using only the portfolio; she misses the step in which open questions are listed separately, so he adds it.
Why this works
The test by a colleague revealed a missing step. This way, the portfolio becomes a reusable working method and not a collection of separate results.
Make it slightly more difficult
Choose the date on which you will look at your portfolio again, and note which application you will test again first.
Extra practice, at your own pace. You do not need to submit anything. This does not count towards an automatic assessment.
Your workplace · learning by doing
Make something. See what changes.
Checking the AI connection… You can already view the written example.
Present your assignment to Ende.
Read the answer, change one thing, and compare.
Previously held
New version
You remain in control
What do you want to improve in your next attempt?
On this device only. AI answers remain only in this page memory. This is self-review, not a grade.
Why this approach works
The test by a colleague revealed a missing step. This way, the portfolio becomes a reusable working method and not a collection of separate results.
A step further
Choose the date on which you will look at your portfolio again, and note which application you will test again first.
Try & discover · no obligation
Put the steps in a workable order
Move the steps with the arrows. Check whether each step has its necessary input and control.
What does a step depend on?
- Choose three applications from your own work must precede Replace confidential data with fictional examples
- Replace confidential data with fictional examples must precede Add your method card for each application
- Replace confidential data with fictional examples must precede Add test sets and their results
- Add your method card for each application must precede Write a reflection for each application
- Add test sets and their results must precede Write a reflection for each application
- Write a reflection for each application must precede Have someone read and try out the portfolio
- Have someone read and try out the portfolio must precede Plan your next learning step and a review
Move a step and then check the sequence. The arrows also work with your keyboard.
This model checks dependencies, not duration, capacity, or actual execution. With multiple agents, the human decision remains a separate step.
This trial works locally with fictitious data and does not count towards your lesson completion. It starts anew with a new lesson or reload.
A prompt to get you started
Help me create a structure for my personal AI practice portfolio. I want to include three fictional applications: a document workflow for invoices, an automation for draft answers and a short explainer video. For each application, give the headings goal and task, result, method card, test set with results, data agreements, tool used and reason, and reflection. Add a final section with my next learning step and a review date. Do not fill in any content I did not give.
Apply it
You want to share your AI practice portfolio in a job application. It contains real customer emails from your current job as examples.
Which approach best fits this situation?
- Keep the emails in; they show best what you can do.
- Replace the real emails with fictional or cleared examples and use them to show your way of working and your checks.
- Remove only the names from the emails and share the portfolio.
With JavaScript, you can check your approach and receive immediate feedback.
Test your insight
What makes an AI practice portfolio most valuable for someone who wants to assess or take over your work?
- A collection of impressive end results without explanation.
- For each application, the method, the checks with their results, the data agreements and an honest reflection.
- A list of all the AI tools you have ever used.
Extra practice
Put together your portfolio with three applications. Have it read by someone who was not involved and ask whether they could repeat one application with only your portfolio.
Application 1: a document workflow for five fictional invoices, with a schema, VAT check and approval by the team leader; the check caught a misplaced comma. Application 2: draft answers for a community centre, with a test set of eight cases, including an injection attempt. Application 3: an explainer video about e-books, with proofread subtitles and a mention of AI use. Reflection: with the video, I trusted the subtitles too quickly. Next step: a test set for my email templates, ready by the end of next month.
