Inspiration
PillBuddy was inspired by personal experience and firsthand observation of the challenges of medication adherence. With daily prescriptions so common, it seems counterintuitive that missing doses remains such a widespread issue. Failure to take medication properly can lead to serious adverse health consequences, especially for individuals living with dementia or Alzheimer’s disease, where memory loss makes adherence even more difficult. We set out to create a product that is simple to use, affordable, and something we would genuinely want our own families to have access to.
What it does
Pill Buddy is an intelligent medication storage and reminder system. Using an ESP32 microcontroller, Pill Buddy detects when users add or remove pill bottles from the device. Once a bottle is detected, it applies Agentic AI from ElevenLabs to identify the prescription and retrieve relevant medication information. The system then maintains a running pill count, tracking how many pills remain as you use it. Users can interact with Pill Buddy through an iOS app built with Expo React Native, and all data is backed by AWS Cloud Services to ensure reliable storage and seamless access across sessions.
How we built it
We used a multitude of technologies spanning from 3D Printing, AWS Cloud Services, an ESP32 Microcontroller, ElevenLabs, an IOS app, and more.
Challenges we ran into
In the process of creating our project we ran into numerous challenges along the way. The first major challenge we faced was in the hardware available to us. We designed it with the intention of using a limit switch to determine if the pill bottle is present in the holder however we were not able to find one to use. To solve this problem we designed a 3d printed housing for a momentary push button that allowed us to turn it into a functional limit switch. We also ran into problems with the Raspberry Pi not being able to boot with the microSD card provided. We solved this problem by utilizing an unfamiliar microcontroller programmed in C with a RTOS and we learned a ton along the way.
Accomplishments that we're proud of
Designed and 3D printed a custom mechanical alternative to a limit switch Complete CAD design of the Pill Buddy system, which includes a model from concept to final assembly and animations for understanding Programmed the ESP32 using embedded C, implementing control logic, I/O interfacing, and system integration Elevenlabs Voice Agent Features Expo React Native Client Application Product-Wide AWS Usage
What we learned
How to quickly iterate and turn 3D models into functional prototypes through rapid testing and real-world validation. How to adapt quickly to design challenges that we did not anticipate and find effective solutions under pressure. How to use Amazon Web Services to implement a full-scale product, including cloud storage, authentication, and serverless functions.
What's next for Pill Buddy
Some additional features that we would like to add in the future include support for medications that need to be taken more than once daily or at less than daily intervals, a noise feature that will audibly remind users to take their medication, and an auto-reorder feature to handle managing medication refills automatically using Amazon Pharmacy.
Built With
- amazon-api-gateway
- amazon-cloudwatch
- amazon-dynamodb
- amazon-web-services
- c
- elevenlabs
- esp32
- expo.io
- lambda
- python
- react
Log in or sign up for Devpost to join the conversation.