How We View AI

“If AI writes the code, why should my kid learn any of this?”

It is the question we get most, and it deserves a straight answer rather than a sales pitch. So here is ours, in one page.

We think the honest answer has two halves. The first: learning to type syntax from memory matters less every year — pretending otherwise would be dishonest, and your child would notice before you did. The second: someone still has to decide what to build, judge whether the result is any good, and fix it when it is not. That work is getting more valuable, not less.

So we teach both. Coding is how a student learns to understand. AI is how they build. A child who only has the second half is stuck trusting whatever the machine hands them.

01

What we believe

The Skill Moved

The valuable part was never the typing.

Remembering syntax is the part AI genuinely replaced. Deciding what to build, breaking a messy problem into parts, and recognising a wrong answer are the parts it did not. Those were always what a good coding class taught — they are simply now the whole point rather than a side effect.

Using ≠ Understanding

Everyone will use AI. Far fewer will understand it.

Your child can already get an answer out of a chatbot without us. What they cannot yet do is tell when that answer is confidently wrong, explain how the thing works, or build something better. That gap is the entire job we have.

Supervision

Unsupervised is the actual risk.

Most students are already using AI for schoolwork, with or without guidance. The realistic choice in front of a parent is not whether their child uses it — it is whether anyone is teaching them to use it honestly and well. We would rather that be a live instructor than nobody.

02

How we teach AI

The Question Every Parent Asks

“Isn't this just teaching my kid to cheat?”

It's the right question to ask. Your child is already using AI for homework — most students are, with or without guidance. The difference is whether anyone is teaching them to use it well. That is a large part of what we do.

We teach where AI is wrong

Students learn to spot confident nonsense: made-up sources, broken logic, code that runs but does the wrong thing. Checking the answer is the skill, not getting one.

Understand it or you did not build it

Students explain their own projects line by line. If AI wrote something they cannot account for, that is where the lesson starts, not where it ends.

The honest line, drawn clearly

Where AI help ends and academic dishonesty begins — citation, disclosure, and school policy — taught explicitly, so your child is not guessing.

A human watching the whole time

Classes are live with a real instructor who sees how your child works, not just what they submit. Unsupervised AI use is the actual risk; this is the answer to it.

Built in Class

What students actually ship with AI

Not demos. Real projects our students designed, built, and can explain line by line.

A study-quiz app

Claude Code · Age 12

Turned a chapter of biology notes into a working quiz app with a score tracker — then rewrote the scoring logic by hand once she understood it.

A homework research agent

Agentic AI · Age 16

An agent that plans its own research steps, searches, and cites its sources — built to show why unsourced AI answers cannot be trusted.

A story-writing partner

ChatGPT · Age 9

A guided prompt tool that helps younger kids draft stories without writing the story for them — his own rule, added after a lesson on academic honesty.

Give Your Child a Head Start on AI

Live online classes where kids ages 6–18 build real apps, games, and AI agents with a real instructor on video — not a chatbot, not a pre-recorded course. Your first class is free.