Foundations: Finishing & Precision
The hardest skill in software is finishing something. The second hardest is getting the details right. At this stage, kids learn both: completing projects from start to finish, while discovering that computers take everything literally.
What they're learning
- Computers do exactly what you tell them. Nothing more, nothing less.
- Projects need a goal, not just "doing stuff"
- Every single character matters. There's no "close enough."
- Error messages are clues, not criticism
Example projects
Does it have a goal, a loop, and an outcome?
Every character matters. Reading error messages is the whole lesson.
A real URL someone can visit. That's what shipping feels like.
Celebrate finished work, not ideas. Don't fix errors for them. Ask "what does the error say?" Kids who learn to ship and debug at this age carry those habits for life.
Problem Solving: Breaking Things Down
A hard problem is just a bunch of easy problems stuck together. At this stage, kids learn to break things into pieces and understand how programs keep track of what's happening.
What they're learning
- Big problems get easier when you split them into smaller parts
- If you solve something once, save it so you don't have to solve it again
- Some programs run forever, always checking what's changed
- Programs remember things (like your score or what level you're on)
Example projects
Movement, collisions, scoring: each gets its own section.
Objects bouncing, things colliding, actions triggering other actions.
What does each piece need to work on its own?
Ask "how would you break that into steps?" in everyday life, not just coding. It's a habit that helps everywhere. This is usually when kids start feeling really proud of what they've built.
Real World: Hardware & Users
Code is inside everything. At this stage, kids connect the digital and physical worlds, and shift from asking "does it work?" to "can someone else actually use this?"
What they're learning
- Code can control things in the real world
- Physical stuff breaks in different ways than software
- Good design is about how something works, not just how it looks
- If a user gets confused, that's on the builder to fix
Example projects
Code meets the physical world.
Strangers can use it without instructions.
Players you'll never meet will try it.
Hardware is frustrating. That's actually the point. Have someone outside the family try what your child built. Watch what happens when they get confused.
Advanced: Big Projects & AI
Big projects fail because of bad organization, not bad code. And in 2026, the question is: is your child directing the AI, or is the AI directing them?
What they're learning
- Planning before building isn't cheating. It's the job.
- Big projects are really just a bunch of smaller pieces working together
- AI can generate code, but it can't decide what's actually worth building
- Knowing when not to trust a tool matters just as much as using it
Example projects
A real product they work on over several weeks.
They're in control of the AI, not the other way around.
The whole thing: what users see, the logic behind it, and putting it online.
Ask your child to explain how their project works without the computer. If they can't, they used the tools. If they can, they directed them. That distinction is everything.
The danger isn't that AI will replace your child. The danger is that your child will let AI replace their thinking. Everything in this roadmap builds that judgment. Don't skip the steps.
What connects all four stages?
Every stage ends with something real. Not a test score, not a grade, not a certificate. Something your child made, that works, that they can share. That habit (finish it, ship it, show it) is the most important thing we teach. Because in the world your child is growing up in, ideas are cheap. Actually building things is everything.
A note on the "AI will replace coders" question
We hear this from parents all the time. Here's what we've seen: AI makes kids who understand how things work way more capable. It makes kids who don't understand way more dependent. The goal of this entire roadmap is to build that understanding. Not so your child can compete with AI, but so they can think with it.