A little learning.
A lot of possibility.
A few practical starting points for curious minds. Try an exercise, make it your own, and keep asking better questions.
Three practical learning exercises
01 / TRY IT NOWA better prompt
in four small steps.
Give your question context, a purpose, a format, and a way to check the result.
Open the prompt lab ↗02 / LEARN THE HABITGood answer?
Let’s check.
A simple routine for reading AI output with a curious, questioning mind.
Read the trust check ↗03 / MAKE SOMETHINGTurn a small problem
into a first project.
Find a useful idea, keep the scope small, and show what you learned.
Try the project brief ↗Make your question
a little clearer.
Build a reusable prompt with this simple template. It runs entirely in your browser; it doesn’t call an AI model.
Explain how machine learning works to a complete beginner. Start with a plain-language explanation, then give one concrete example and a small exercise I can try. Define technical terms and suggest trustworthy sources I can check.
Use an age-appropriate tool. Don’t include private information. Always check the result.Confidence isn’t
the same as accuracy.
AI can produce convincing text that gets things wrong. Treat an answer as a starting point for investigation.
- Separate claims from suggestions.
Highlight statements that can be checked: dates, quantities, names, sources, or technical steps.
- Go to the original source.
Open the source yourself. Check that it exists, is relevant, and actually supports the claim.
- Ask what is missing.
Look for uncertainty, limitations, alternative explanations, and whose perspective is absent.
- Know when to ask a person.
For consequential decisions, involve a qualified person rather than relying on an AI response alone.
Small enough to finish.
Useful enough to matter.
Choose a familiar, low-stakes problem. A reading planner, an illustrated story, or a public-data research journal is a better first project than an everything-app.
- Write the brief in one sentence.
“I want to help [someone] do [one thing] more easily.”
- Choose one useful output.
A short plan, a working flow, a page, or a simple prototype. Leave the extra features for later.
- Try three different inputs.
Include an ordinary case, a confusing case, and one your project should refuse or flag.
- Show the process.
Share what you made, what AI did, what you checked, and what you would change next.
Your first project
could start here.
Explore the learning stage that fits what you want to make.