Toddler Edition · Ages 18 months – 4 years

Your child's brain is already doing the work.

Most toddler apps skip the most important step: teaching your child that the screen is real. Training Minds starts there — then delivers the same thing that makes machine learning work: clean, categorized signal. No engagement algorithms. No data collection. Every setting is yours.

No data collected or shared No engagement algorithms Works offline No toddler-facing progress pressure No ads, ever
Get notified at launch → Via Substack. Unsubscribe anytime.
Coming to:
🍎 App Store
Google Play
45
lessons across 10 domains
1,500+
professional voice clips
600+
photorealistic images
2
languages, separate progress
0
bytes of data collected
The science

The prerequisite most apps skip.

Before a toddler can learn from a screen, they need to understand that the screen represents reality — that the dog on the screen is a dog. This gap, part of what researchers call the screen deficit, affects children under 30 months most acutely. Training Minds is the only toddler learning app that addresses it directly, before any lesson runs.

"The screen deficit isn't a mystery — it's a calibration problem. Solve the calibration, and the deficit mostly disappears."
1

Live Mirror

The child sees themselves on screen in real time — the simplest, most direct bridge. What happens to me, happens on the screen.

2

Self Video

Familiar routines played back. The child sees that the screen can contain events they've actually lived — building representational confidence.

3

Face Detection

A red dot appears on your child's nose. Or ear. Or chin. They tap it, the voice names it, and it moves somewhere new. Simple, a little silly, and surprisingly hard to put down.

4

Familiar Faces

"Where's Grandma?" Recognition using the child's own family photos, stored only on the device. They never leave it.

5

Favorite Things

Their actual toys and objects on screen. The final bridge: if my bear can be on the screen, so can a zebra.

Learning isn't easy — but it can be more efficient.

Toddlers learn through experience. The brain is extraordinary at extracting patterns without explicit teaching. But that process takes time, repetition, and enormous amounts of noisy, unstructured signal. Think of it like training a machine learning model on raw, unlabeled data — given enough time and a few cities' worth of data centers it will learn — but clean categorized signal gets you there faster. This app doesn't replace the world. It cleans the signal.

"The same brain that would eventually learn 'dog' from a thousand ambient encounters learns it faster and more reliably when the signal is clear, varied, and well-timed."
Seed

Generalized exposure

Seeding happens both on and off the screen — real-world exposure matters most. Foam bath letters, resin animal figurines, picture books. The app's Seed mode delivers clean, label-only exposure for earlier lessons, meeting toddlers where they are as they learn to learn from a screen.

Train

Contrast and labels

When your toddler shows interest in naming and categorizing things, contrast is one of the most effective teachers — "that's a horse, this is a zebra." The app builds those contrasts into recognition games. Not every type of learning benefits from this; we apply it selectively.

Deploy

Your job in the real world

You bring the learning off the screen. Point to the zebra at the zoo. Hand them the foam letter. The app prepares the brain — you're the one who deploys it, in the park, at the grocery store, everywhere.

Features

Built for toddlers. Controlled by parents.

Every design decision is backed by research — and every decision that was ours to make is adjustable. Because no two children are the same.

Click any tile to see it in the app →

🔒

No data, ever

No account. No cloud sync. No analytics. Everything stays on your device. Transfer to a new device with a 6-digit code — nothing ever touches our server.

🔬

Every algorithm documented

Lesson order, mastery rules, how quickly new items rotate in — all explained in plain language inside the app. We've created defaults based on research, but none of them are locked.

✈️

Works completely offline

After the initial download, no internet connection is needed. Plane, road trip, anywhere. All 1,500+ audio clips and 600 images are pre-loaded and cached locally.

🎨

Two visual styles

Bold tiles with vivid high-contrast colors, or Natural Wood with warm Montessori-inspired tones. Switch anytime.

🌎

English + Spanish

1,500+ professional voice clips per language, recorded separately — not translated. Bilingual children get completely independent progress tracking per language. Switch with a few taps.

👨‍👩‍👧

Multiple child profiles

Each child gets their own profile, their own progress, and their own lesson path. Switch between children with one tap — everything is kept completely separate.

🌍

Built for every family

Our lesson images reflect the real, diverse world. And you decide what fits yours — hide any item that doesn't match your family's values, dietary choices, or cultural context. The defaults are inclusive; the controls are yours.

🎛️

45 lessons — you're in control

45 lessons across 10 domains. We've built algorithms to show your child what they're ready for and make lessons trickier as they master them — but you can hide any lesson, skip any prerequisite, or adjust the difficulty yourself.

📊

Calm progress tracking

No scores, no rankings, no percentiles. Just a clear, readable picture of what your child is playing and how they are progressing.

About

The ideas behind the app.

Published on Substack. Free to read.

Screen calibration
The video deficit isn't what you think it is.
Why toddlers learn less from screens — and exactly what fixes it.
Read →
Language learning
The foam letter story.
How a bathtub full of letters taught me everything about how toddlers build categories.
Read →
Letters
Letters as primitives.
Why the alphabet should be taught as shapes before it's taught as sounds.
Read →
Training Minds

A book is forthcoming.

I'm a mom of a toddler. I also spent my career in data and analytics — and when I started researching what my son actually needed from a learning app, I couldn't find it. So I built it. The frameworks behind this app — first Screen Calibration, then Seed, Train, Deploy — are the subject of a book I'm writing. This app puts what I've learned into your hands now.

Amazon · coming soon Barnes & Noble · coming soon Apple Books · coming soon