Homeschool parents ask us this every September. We have a seven-year-old, a ten-year-old and a fourteen-year-old. What do we do for technology, and how do we sequence it so we are not buying twelve different subscriptions every year? Most existing answers come from platforms trying to sell their own boxes. This one is what we actually teach across all of our programs, written so a family with two laptops and a normal weekly schedule can follow it.
The short version is this. There is a clean four-rung ladder from age seven to eighteen. Each rung lasts roughly three years. Each rung has a free path and a paid path. The job is to know which rung the child is on, finish that rung properly, and move on. Skipping rungs causes the dropouts we see most often. Lingering too long on early rungs causes the boredom we see second most.
Why a 12-year plan beats one-off courses
Homeschool families are buying tech education in a market that has changed twice in five years. The World Economic Forum's Future of Jobs Report 2025 ranks AI and big data, technological literacy, and creative thinking as the three skills with the fastest growing demand through 2030. None of those is a single course. They are layered competences built over years. A one-off coding camp is a starting point, not a strategy.
A family that plans the full ladder makes better decisions every September. They know when to invest in a real cohort and when a free tutorial is enough. They know when to add a paid mentor and when to back off. The savings compound.
Rung 1. Ages 5 to 10: Scratch and the joy of making
The first rung is Scratch from MIT. It runs in any browser. It is free forever. It has been used by millions of children since 2007 and the platform is mature, well-translated, and safe.
The goal at this age is not to ship a product to the public. The goal is to build the mental model "I can make the screen do what I tell it." That model is the foundation for everything else.
A reasonable Scratch year looks like this. Two short sessions a week, twenty to thirty minutes each. Build small things. Animated stories. Two-button games. A birthday card for a grandparent. By the end of the year a seven-year-old should be comfortable with sequences, loops, events, and one or two variables. By the end of three years a ten-year-old should be able to build a multi-level game with custom logic, broadcasts, and lists.
You do not need a paid subscription. The Scratch website is enough. If you want structure, the code.org K-5 curriculum is free and well-aligned. If you want a cohort experience to motivate a kid who learns better with peers, the Summer Coding Bootcamp Logic Builders track is built precisely for this rung.
Two practical rules at this stage. First, no text-based code yet. The cognitive jump from blocks to typing real syntax lands much better closer to age ten or eleven. Second, the child has to demonstrate one Scratch game with their own custom logic before they move up. Not a copy of a tutorial.
Rung 2. Ages 9 to 13: App Inventor and the first real app
The second rung is MIT App Inventor. It is the closest thing to magic in the entire stack. MIT reports the platform has over 900,000 unique monthly active users from 195 countries and has been used to create more than 34 million apps. It is free.
What App Inventor unlocks is the first moment a child runs an app they made on a real phone. I have watched that moment hundreds of times. It changes the kid. The screen stops being a place adults work and starts being a place they can build.
Plan the year around two ideas. First, deepen logic. Variables, lists, procedures, multiple screens, the basic vocabulary of real programs. Second, ship to one real user. Not a parent. A cousin. A friend at church. A classmate. The feedback from a real user is the most valuable teaching aid you can buy and it costs nothing.
This is also the rung where you start introducing AI as a thinking partner rather than a homework crutch. We will get to that in a minute.
The free path is the App Inventor website plus the official tutorials. The structured path is our Summer Bootcamp Code Masters track. The cohort experience matters more at this rung than at Scratch because the work gets harder and motivation dips midway.
A child has finished this rung when they have shipped two real apps that one real user has installed and given feedback on.
Rung 3. Ages 12 to 15: Python, AI literacy, and the read-debug-direct habit
The third rung is where things get strategic. Two skills come in at once. Real text-based code, usually Python or JavaScript. And serious AI literacy.
For text-based code, the right starting point is Python. It is the working language of AI tooling, automation, and data work. The free path is the Python.org beginner tutorial plus a project-led path like building a small command-line game, then a small web scraper, then a tiny notebook that calls an LLM. Spend less time on syntax and more time on small projects that work end to end.
For AI literacy, the Common Sense Media report The Dawn of the AI Era found that 70 percent of teens have already used generative AI, but only 37 percent of parents whose teen used it knew about it. Half of parents had never discussed AI with their child. That gap is dangerous and it is also fixable.
At this rung your child needs three habits. Read code the AI produces and identify what it does. Debug calmly when the model is wrong. Direct AI with structured prompts rather than vague requests. We dedicate full modules to each of these in the AI Developer track and they are the most often-cited reasons families say their kids became different builders after the course.
Two non-negotiable household rules go in here. AI use is supervised in shared spaces, not in bedrooms. And AI use is reported back. The child tells you what they used AI for that week. Not as punishment. As a habit. The conversation is the thing.
Rung 4. Ages 14 to 18: Idea2MVP and shipping for real users
The fourth rung is the one most families do not realize is available to homeschoolers. This is where a teenager learns to take an idea and turn it into a working product on a live URL that real people use.
The market has changed. Lovable hit 400 million dollars in annualized revenue in February 2026 and reports over 25 million projects created. AI app builders went from toys to production tools in eighteen months. A fourteen-year-old today can ship what a thirty-year-old needed a team of three to ship in 2022.
The skills at this rung are different from the lower rungs. Problem selection, scope discipline, prompt engineering, AI-builder fluency across Lovable, Bolt.new, v0 and Replit Agent, lightweight database work with Supabase, basic analytics with PostHog, talking to real users and incorporating feedback.
Our Idea2MVP course is exactly this rung packaged for homeschool families. Ten weeks. Free-tier tools end to end. A live URL by week five. A public Demo Day at the end. Capstone artifacts that double as university-application material.
A teenager finishes this rung when they have a live URL, five real users who are not family, a demo video, and a one-page write-up of what they learned. That packet is more impressive in 2026 admissions and internship conversations than a transcript alone. We have watched it happen.
What about Code Prodigy and the elite path
A small number of homeschool families have a child who has already finished Idea2MVP at fifteen and wants more. Code Prodigy is our application-only track for that. It runs longer, costs more, and pairs the student with industry mentors. It is not for most families. If your kid is asking for it by name, you will know.
The weekly rhythm that actually sticks
Homeschool tech plans usually fail not from a bad curriculum but from a schedule the family cannot sustain. Here is the rhythm we recommend.
Two sessions a week of focused work. Twenty to thirty minutes at Rung 1. Forty-five to sixty minutes at Rung 2 and 3. Ninety to a hundred and twenty minutes at Rung 4. One short Friday "show me what you made" conversation per child. One monthly "show one real person" requirement. One quarterly project review where the parent and child look at what was shipped, what stalled, and what to start next.
The biggest predictor of success is not raw hours. It is the show-and-tell ritual. The child who has to demonstrate their work to a non-parent every month builds three times faster than the one who does not.
Sibling math and the multi-child plan
Almost every homeschool family I talk to has more than one kid. The strategy that works is to put each child at their own rung simultaneously, sharing one device when budget is tight, and use the older sibling as a junior assistant for the younger one for ten minutes a week. The teaching the older kid does is the most valuable form of practice for them. The mentoring is genuinely useful for the younger kid. Both sides benefit.
Sibling discounts in our cohorts stack precisely so multi-child families can scale this plan without the cost doubling.
What this means strategically
The labor market the WEF report describes is one where the gap between average and capable widens every year. A homeschool family with a deliberate twelve-year tech plan is putting a child in the capable bucket by default. The plan does not have to be ours. It does have to be real, sequenced, and reviewed.
If you want to start where your child is right now, the free assessment class is a 1:1 evaluation that lands a child at the correct rung without guesswork. If you already know which rung they are on, register straight into the relevant program. Sibling pricing computes live. Scholarships are available for families that need them.
The kids in our cohorts who do this from age seven walk into university at eighteen with a portfolio that takes the conversation past the transcript. That is not a marketing line. That is the consistent shape of the homeschool families who plan the ladder out.
Sources