Anthropic just announced it will spend $150 million placing 1,000 early-career workers inside nonprofits at $85,000 a year to teach them how to use AI.
The program is called Claude Corps, and the reason it exists is not complicated: AI is displacing jobs faster than the workforce can absorb the change, and someone decided to write a check instead of a think piece about it. That's the world your kids are growing up in.
The Claude Corps fellows are 18 to 22 years old. They are the last cohort for whom this kind of scramble is even possible, catching up on AI literacy in their early 20s and still finding work on the other side. The kids who are 4, 8, and 13 right now are not getting a fellowship. They are getting a job market that will assume AI fluency the way our generation assumed email.
My son is a baby. I think about this constantly.
So I built this list. Not the generic "coding for kids" roundup you've seen a hundred times. These are the programs I'd actually put my own child in, organized by age, by what they're actually teaching, and by what the evidence says works. I've gone through every one of them.
Start somewhere this week. One program. One conversation with your kid. That's the whole ask.
AGES 4–9 · THE FOUNDATION Where logic and computational thinking begin disguised as play.
01. Scratch — Ages 5+ MIT-built. Kids create real games and animations using logic blocks — genuine computational thinking, not dumbed-down coding. This is where Roman starts. Browser-based · Free · International scratch.mit.edu
02. Code.org — Ages 4–18 The entry point for most kids. Structured tracks from kindergarten through AP-level CS. Works in any browser, no setup required. Browser-based · Free · 45+ languages code.org
03. AI4K12 — Ages 5–18 The only K-12 framework built specifically around AI concepts, not just coding. A great primer for parents too. Self-guided · Free · US, adaptable globally ai4k12.org
04. Kodu Game Lab — Ages 8+ Microsoft-built. Kids create 3D games using visual programming. It feels like play but builds real logic and systems thinking underneath. Windows / Xbox · Free · International kodugamelab.com
AGES 7–12 · BUILDING REAL THINGS Where kids stop consuming technology and start making it.
05. iD Tech — Ages 7–19 Camps and online courses covering Python, machine learning, and robotics. The peer environment is the point — kids who build alongside kids who care, that compounds. In-person + online · $500–$2,000+ · US campuses + online idtech.com
06. Machine Learning for Kids — Ages 10+ Kids train actual AI models — image recognition, text classifiers, real outputs. It demystifies AI at exactly the age when most kids think it's magic. Browser-based · Free · International machinelearningforkids.co.uk
07. MIT App Inventor — Ages 10+ Kids build real Android apps from a browser with no coding background required. I've seen 12-year-olds build things with this that adults couldn't. Browser-based · Free · International appinventor.mit.edu
AGES 13+ · THE SERIOUS TIER For the teenager ready to go from learning AI to building with it.
08. Inspirit AI — Ages 13–18 Built by Stanford and MIT alumni for high schoolers. Project-based AI work in healthcare, climate, and social impact. This is the one I'd put a serious teenager in. Online + in-person · $800–$1,500 · Online global, in-person US inspiritai.com
09. Girls Who Code — Ages 13–17 Free clubs and summer intensives for girls. The confidence gap in tech starts young. This closes it early. Clubs + virtual · Free · US clubs, online global girlswhocode.com
10. fast.ai — Ages 15+ A motivated 16-year-old who finishes Practical Deep Learning will know more than most college grads. Free and world-class. I'm not exaggerating. Self-paced · Free · International fast.ai
The most valuable thing you'll ever invest in isn't a company. It's your kid. Pick one program this week. Watch what happens.
Hit reply and tell me: how old are your kids, and what are you already doing? I read every response.
Jenny
P.S. If you found this useful, forward it to a parent in your life who is thinking about this. The earlier they start, the wider the gap.
