GrowMyBaby
- Role —
- IoT & ML Capstone Developer
- Date —
- June 30, 2024
- Stack —
- Python, Streamlit, Machine Learning, IoT, Groq API
GrowMyBaby is an integrated health-monitoring system developed as our capstone project for Samsung Innovation Campus Batch 5 (in partnership with Skilvul). The platform combines physical IoT hardware sensors with machine learning to track early childhood growth metrics and detect signs of stunting.
Project Overview & Architecture
The primary goal of GrowMyBaby is to simplify baby health tracking for parents and health workers:
- IoT Sensor Measurement: Physical sensors measure and track a baby’s height and weight metrics in real time.
- API Gateway Transmission: Collected hardware data streams through an API gateway to a web application.
- ML Growth Classification: Machine learning algorithms evaluate the incoming growth data against standard health indicators to flag potential stunted growth.
- AI-Powered Assistance: An interactive chatbot powered by Groq LLM APIs offers parents tailored recommendations and health advice based on their baby’s status.
My Core Technical Contributions
- Machine Learning Pipeline: Evaluated, selected, and trained classification models to analyze height/weight parameters and accurately categorize stunting risk levels.
- LLM & Chatbot Integration: Tailored prompt templates and integrated Groq’s high-speed AI Inference API inside a Streamlit web interface to facilitate interactive user support.
- Rapid Problem Solving: Adapted quickly to tight project deadlines, picking up Python ML concepts and API integration patterns on the fly.
Project Links & Deliverables
- 🌐 Live Web Application: GrowMyBaby Website
- 💻 Source Code: GitHub Repository
- 📊 Pitch Deck: Presentation Slides
- 🎥 Video Demo: Presentation Recording