Students working on real problems in our community.

Projects

Five problems, and what we are doing about them

1

1

NevMiss

Web · nevmiss.com

The problem. An older person who does not speak English, or someone living with dementia, can get lost and have no way to tell a stranger who they are or how to reach their family.

The idea. A family registers a parent or a pet and gets a QR tag for them to carry. Anyone who finds them can scan it with a phone camera and reach the family. The family chooses how: a direct call, or a message that keeps their phone number hidden.

So far. Sophia built it herself, from the idea to the working site, using AI coding tools. Several hundred registrations so far, after she and her classmates introduced it at senior centers and animal shelters.

Visit nevmiss.com

2

2

SpeakBit

Web prototype · award winner

The problem. A student who is still learning English can understand the subject and still lose the thread of a fast lecture. One missed explanation makes the next lesson harder, and that is true for any student.

The idea. Record a class, turn it into text, and let AI compress the whole lecture into a review a student can read in a few minutes.

So far. The idea is Kaili Lin's, a classmate, and she leads the project; Sophia built the prototype with her. It received the Best Problem Insight Award at the Sunshine Youth Entrepreneurs Challenge 2026 in Orange County, California, and is now being introduced at school.

Read the news coverage

3

3

PrincessMath

Web game · kindergarten and grade 1

The problem. Sophia's younger sister was starting kindergarten math. Worksheets do not hold a child that age for long, and many apps cost money or are full of ads.

The idea. A kingdom of 30 short levels, from a meadow to a lake to a castle garden. Each level is ten questions that bring one lost animal home. A wrong answer gets a hint and another try. Questions are read aloud, and a parent can record the praise in their own voice. In English and Chinese.

So far. The idea and the design decisions are Sophia's; the game was built with AI tools. Free in any browser, with no ads and no sign-up. More than 40 children have played it so far. What worked here became the starting point for Countiva.

Play PrincessMath

4

4

Countiva

iPhone and iPad app · coming soon

The problem. Many elementary students never get enough practice to feel sure of themselves in math, and practice that feels like a chore does not get done.

The idea. Math practice for US grades 1 to 6 that children choose to play, in English, Spanish and Chinese. No scores to chase, no timers, no ads and no tracking. Grades 1 and 2 are free, for good. So are the first 10 levels of Grades 3 to 6.

So far. The idea is Sophia's, and so are the product decisions: what each grade covers, how a level plays, and that there are no scores or timers. She wrote many of the prompts for the AI tools that built it and made the illustrations with ChatGPT. Version 1.0 has been submitted to the App Store.

About Countiva

5

5

Tally

Web prototype · in progress

The problem. Prices keep rising, and many families nearby feel it. They know where the big money goes, but not whether the phone, the internet or the groceries could cost less. Many also carry credit card debt, and the interest is heavy.

The idea. One simple app that helps a family spend less: a better phone or internet plan, a cheaper way out of card debt, the store nearby with the lower price, and bookkeeping easy enough to do by voice.

So far. A prototype covers phone plans and credit card interest. The rest is still being built, and no family is using it yet.

About Tally

How the work goes

An idea is the easy part

1. Start from a real personEvery project begins with something one of us saw happen to people around us, not with a technology looking for a use.
2. Ask who else has itOne person's problem usually belongs to many. One lost elderly person becomes anyone who cannot explain who they are; one confused classmate becomes any student who falls a lesson behind.
3. Build, fail, come backThe first version of NevMiss did not work and the idea sat unfinished. AI coding tools made it possible to return to it: break the idea into small problems, test what comes back, revise.
4. Take it to peopleA tool nobody uses helps nobody. Getting people at senior centers, shelters and school to understand and try something is a lesson of its own.

Still unfinished: sturdier NevMiss tags, getting SpeakBit into real classrooms, Countiva's first release, and Tally itself.

Who is behind this

Cyanworld is led by students