Standalone lesson.
Lesson 1570 of 1570
Your First Capstone
Design an AI assistant that helps your community.
Time to build something. Your Builder capstone asks you to design an AI assistant that helps your community. You’re not coding it yet — you’re designing it. That’s the hardest part anyway.
The four questions
- Who is it for?Be specific. “My grandfather who’s learning to use email.” Not “old people.”
- What does it help them do? One thing. Not ten.
- What would a great first conversation look like? Write it out like a script.
- What’s one way it could go wrong, and how would you prevent that?
Examples from past capstones
- A homework buddy that won’t give you the answer — only hints — for kids whose parents can’t help with math.
- A pet-care advisor that warns you when a home remedy is actually unsafe for dogs.
- A library chatbot that suggests the next book based on the mood you’re in, not just the last book you read.
- A weekly review assistant for a middle-school student with ADHD that gently prompts them through their planner.
What to submit
Upload a 1-page PDF or image. The Who / What / Script / Failure Mode sections. That’s it. Real, specific, humane. Your librarian (or parent) can review and award the Builder badge.
Why this is the capstone
Designing an AI is not the same as using an AI. This is the practice that matters most as you become a Creator. The code is the easy part. Knowing what’s worth building is the hard one.
Tutor
Curious about “Your First Capstone”?
Ask anything about this lesson. I’ll answer using just what you’re reading — short, friendly, grounded.
Progress saved locally in this browser. Sign in to sync across devices.
Related lessons
Keep going
Builders · 30 min
Why Agents Fail (and How to Notice)
Agents fail in weird, quiet, expensive ways. Learn the six failure modes, the warning signs, and the simple habits that catch problems before they compound.
Builders · 40 min
Builder Capstone: Design an Agent for Your Life
No code. Just design. Pick a real task you do every week and draft a complete agent spec — goal, tools, loop, stop, approvals, and what success looks like.
Creators · 50 min
Evaluating Agent Performance: SWE-bench, WebArena, GAIA
Numbers on leaderboards are seductive and often wrong. Learn the big benchmarks, their leaderboard positions, their recently-exposed cheats, and how to run your own evals.
