Love, Maths

Love, Maths provides preschoolers with one purposeful maths lesson at a time. The lesson are intentionally calm but engaging for purposeful practice rather than endless digital babysitting.

Comment

Love, Maths

Inspiration

Love, Maths was inspired by parenting my elder daughter, Lovette.

My wife and I have always been very cautious about screen time and its potential detrimental effects on young children, such as developmental delays. Lovette currently has very limited screen time, mainly during holidays and our monthly family movie night.

However, as she approaches primary-school age, we started to become concerned and worried about whether our personal parenting choices would put her at a disadvantage in this digital age and leave her unfamiliar with digital experiences.

Henceforth, I began exploring whether technology could be introduced safely in a way that supports purposeful learning without relying on overstimulation, endless content or addictive reward loops. More importantly, I want to bring control over the digital experience back to parents.

And voilà, Love, Maths was created with the following overarching design concept:

Can a children’s learning app feel engaging while still being calm, purposeful and easy to stop?

What It Does

Love, Maths is a calm, animated toy table for early mathematics.

Children aged approximately two to three participate in short Play Together activities with an adult, while children aged approximately four to six gradually move towards independent, content-bounded practice.

Softly rendered 3D objects intuitively invite children to touch, tap and move them. Important mathematical content responds visually when selected, and an original unnamed baby-quokka helper supports selected transitions between activities.

Every activity is designed to remain usable without sound. Child mode contains:

  • No adverts
  • No purchases
  • No external links
  • No shop
  • No infinite play
  • No manipulative streaks or reward loops

Each session focuses on one main mathematical concept and ends naturally once the activity is completed. This aims to reduce a child’s potential tantrum (which is also emotionally draining on the parent) when the digital experience is stopped. Instead of the parent having to interrupt an endless game midway, the app reaches a clear and natural stopping point by itself.

For example, children may explore how two smaller groups can combine to make five (aka number bonding):

[ 2 + 3 = 5 ]

The app presents this first through objects and movement rather than expecting young children to understand written equations immediately.

How I Built It

I developed the product concept by using ChatGPT as my product manager, research partner and plain-English technical guide.

I had brainstormed my broad ideas and intended direction for the app with ChatGPT, which then helped me further refine the direction based on best practices, early-years learning principles and child-safety considerations.

Subsequently, I requested ChatGPT to produce detailed build instructions to be used in Codex, which turned the product requirements into a working prototype.

As I am a solo vibe coder without any coding knowledge, I relied heavily on ChatGPT to help me:

  • Test and challenge the product concept
  • Define the overall product experience
  • Translate parenting concerns into design requirements
  • Review child-safety and usability decisions
  • Structure the development roadmap
  • Explain technical concepts in non-technical language
  • Prepare detailed instructions for Codex
  • Identify gaps and inconsistencies in the implementation (aka playing devil’s advocate)

As mentioned, I had Codex do the coding, while ChatGPT helped me understand what was being built, why it was being built and whether the implementation remained aligned with my original product vision.

Challenges I Ran Into

The biggest challenge was positioning Love, Maths as both engaging and intentionally low-stimulation (boring but not boring).

The product needed to be interesting enough for children to interact with, while avoiding the flashing rewards, endless levels and rapid feedback commonly associated with children’s apps.

In simple terms, it needed to be:

Calm without becoming boring.

Another important consideration was at the parent’s level, as parents are ultimately the decision-makers. The app needs to be “interesting” and meaningful for the parents so that they would be willing to cascade the experience down to their preschoolers (aka the target audience).

Together with ChatGPT, the design concept evolved from a narrow “number bonding up to 10” activity into 2-stage activities intended to suit the different age groups.

Younger children require adult-supported interaction through the Play Together activities, while older preschool children should gradually be able to complete activities more independently.

Another major challenge was more personal — that is, building the product as a solo, non-tech person.

More often than not, I struggled to understand the code and technical architecture. Fortunately, ChatGPT made the process easier by explaining the logic and workflow in non-tech jargon, while Codex translated the detailed product requirements into a working prototype.

Accomplishments I Am Proud Of

Being a non-tech vibe coder, I am proud of developing a coherent product concept centred on child safety, parent trust, purposeful screen time and bringing control over digital experiences back to parents.

Rather than treating safety as an additional feature or an afterthought, it is the centrepiece of the entire product:

  • Sessions are short and content-bounded
  • Child mode contains no commercial content
  • Activities remain usable without sound
  • Error feedback is calm and non-aggressive (aka no big red cross or harsh buzzer sound)
  • Parent reports are protected behind biometrics or a Parent PIN
  • The app does not encourage endless replay
  • Learning content is informed by research and international early-years benchmarks, such as the Nurturing Early Learners Framework from Singapore’s Ministry of Education
  • Shapes, patterns and numbers are treated as distinct learning journeys
  • Artwork, wording, characters and interactions are designed to be original

I am also proud that the prototype is not merely a technical proof of concept.

It demonstrates a complete household journey: the child selects a profile, completes a short activity, reaches a natural stopping point and returns the device to the parent, who can then unlock and review the progress report.

What I Learned

The key takeaway was that AI-assisted development is not simply about asking an AI to generate code.

For the process to work, I needed a clear product vision and concept, detailed requirements and repeated reviews of both the design and implementation.

While ChatGPT helped turn my broad ideas into a detailed build blueprint and Codex helped to build the prototype, human intervention (aka myself) remained imperative throughout the process.

ChatGPT and Codex acted as my trusted partners, not as replacements for human decision-making.

I learned that people without traditional software-development backgrounds can build meaningful applications as long as they are willing to learn, test their assumptions and communicate clearly with AI tools.

I also learned that designing a calm children’s product can be harder than designing a highly stimulating one. Every animation, sound, prompt and reward needs to justify its place. Otherwise, the product may gradually become the exact type of children’s app that I was trying to avoid.

What’s Next for Love, Maths

The next step is to complete and launch the prototype for real-world testing and refinement.

Initial testing will focus on whether:

  • Children understand what to do without excessive adult explanation
  • The touch and movement interactions feel intuitive
  • Activities remain engaging without overstimulation
  • Children accept the natural end of each session
  • Parents understand and value the progress reports
  • The experience works across different preschool ages and abilities

Feedback will be collected through the protected parent area, not at the child's learning layer.

After validating the pilot test, I plan to expand Love, Maths into a structured sequence of early maths activities while retaining the same calm, child-safe and parent-trusted design principles.

Built With

Share this project:

Updates