BabyCare Copilot

Inspiration

As a new parent, I found that baby care information is often scattered across notes, messages, and different apps. Feeding, sleep, diaper changes, and growth records are easy to forget, especially when parents are tired.

This inspired me to build BabyCare Copilot, an AI-powered baby care assistant that helps parents record daily care activities and understand their baby's routine more clearly.

What It Does

BabyCare Copilot allows parents to record:

Feeding time, amount, and feeding type
Sleep start and end times
Diaper changes
Baby weight and growth information
Important daily notes

The application organizes these records into a clear timeline and generates daily summaries.

OpenAI is used to turn structured records into easy-to-understand insights, such as feeding frequency, sleep patterns, and notable changes. These insights are for general parenting support only and are not medical diagnoses.

How We Built It

The frontend was built with UniApp, allowing the project to run as a WeChat Mini Program and mobile application.

The backend was built with Java and Spring Boot, with MySQL used to store baby profiles and daily care records.

We used the OpenAI API to generate natural-language summaries from structured baby-care data. Codex assisted with interface development, database design, API implementation, debugging, and code optimization.

Challenges

One major challenge was ensuring that AI-generated summaries remained clear, useful, and responsible. The application must avoid presenting general observations as medical advice.

Another challenge was converting many small daily records into summaries that are simple enough for tired parents to understand quickly.

We also needed to design an interface that allows parents to complete a record in only a few taps.

What We Learned

Through this project, we learned how to combine structured data with generative AI instead of asking the model to make unsupported guesses.

We also learned how Codex can support the full development workflow, including planning features, generating code, fixing bugs, improving prompts, and writing documentation.