This project leverages a Multi-Agent AI architecture to revolutionize elderly care by combining the power of Large Language Models (LLMs), LangGraph, Pydantic AI, and intelligent AI agents. Each agent is specialized for a distinct care function—such as medication reminders, emotional support, health monitoring, and daily task assistance—and they work collaboratively in a modular, graph-based framework to provide personalized and proactive support for elderly individuals.
- Personalized Elder Care: Modular agents tailored for individual needs.
- Proactive Assistance: Anticipates and responds to daily care requirements.
- Graph-Based Architecture: Ensures smooth and logical coordination between agents.
- Scalable & Extendable: Easily add or update care modules as needed.
-
🤖 LLMs
Powering conversational understanding and decision-making. -
🧠 LangGraph
Orchestrating agent interactions through dynamic graph logic. -
🧩 Pydantic AI
Structuring agent input/output with strong type safety and validation. -
👥 Multi-Agent Coordination
Specialized AI roles working in harmony for holistic elderly care.
- 💊 Medication reminders
- ❤️ Emotional companionship
- 🩺 Health monitoring alerts
- 📅 Daily task scheduling and assistance
Follow these steps to set up and run the project locally with Streamlit:
# 1. Clone the Repository
git clone https://github.com/mukul74/CompanionAI.git
# 2. Navigate to the Project Directory
cd CompanionAI
# 3. Create a Virtual Environment and Install Dependencies
# For Linux/Mac
python -m venv venv
source venv/bin/activate
# For Windows
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt
# 4. Run the Streamlit Application
streamlit run app.py
# 5. From the repo select the test_patients_data.csv in the webpageContributions are welcome! Feel free to open issues, fork the repo, or submit pull requests to improve this system.
Empowering the elderly with intelligent, compassionate, and proactive AI support.
---