HyphochondriaAI - Health Anxiety Specialist

Introduction

HyphochondriaAI is a chatbot designed to provide compassionate support and evidence-based information for individuals experiencing health anxiety. Built with modern web technologies and powered by advanced language models, it offers a safe space for users to explore their health concerns while promoting healthy coping mechanisms and encouraging appropriate medical consultation when necessary.

Technology Stack & Features

• ⚡ FastAPI for the Python backend API

  • 🧰 SQLModel for database interactions (ORM)
  • 🔍 Pydantic for data validation and settings management
  • ✅ Pytest for comprehensive testing
  • 🔄 Alembic for database migrations
  • 💾 PostgreSQL as the SQL database
  • 🔠 Ruff and Black for linting and formatting

• 🚀 React for the frontend

  • 💃 Using TypeScript, hooks, Vite, and other parts of a modern frontend stack.
  • 🎨 TailwindCSS for responsive design
  • 📱 An automatically generated frontend client
  • 🐕‍🦺 Husky with ESLint as Git hook

• 🤖 LangChain for LLM framework

  • 🌐 LangGraph for agent creation and conversation memory
  • ☁️ AWS Bedrock for AI model access
  • 🦾 Claude as the AI model

• 🏭 CI/CD pipeline with GitHub Actions.

• 🐋 Docker Compose for development and production.

Main Page

Conversation Page

Interactive documentation

AWS Cloud Architecture

• ☁️ ECS with Fargate for serverless container orchestration
  • 🐳 ECR for Docker image registry
  • 🔐 Secrets Manager for secure credential storage
  • 📊 Cloudwatch for monitoring and logging
  • 🔄 Auto-scaling and health checks with circuit breaker deployment
• 🌐 CloudFront global CDN for frontend delivery
  • 🪣 S3 for static website hosting
  • 🔒 Origin Access Control (OAC) for secure bucket access
  • ⚡ Custom caching policies for optimal performance
• 📡 Amazon VPC - Custom virtual private cloud with public/private subnets
  • 🏗️ ALB for HTTP/HTTPS traffic distribution
  • 🔓 VPC Endpoints - Private connectivity to AWS services:
  • 🔐 ACM for SSL/TLS certificates
  • 🛡️ Security groups and network isolation
• 🗄️ RDS managed database
  • 🔄 Multi-AZ deployment for high availability
  • 📈 Performance Insights and enhanced monitoring
  • 🔐 Encrypted storage and automated backups
• 🤖 Terraform for Infrastructure as Code
  • 📝 S3 backend with DynamoDB state locking
  • 🚀 Automated deployment scripts

ECS Cluster Metrics

ECS Log Events

RDS Metrics

ALB Metrics

Cloudwatch Alarms

Installation

Follow these steps to set up and run the application locally:

1. Environment Setup

# Ensure you have Python 3.12+ and Node.js installed
# PostgreSQL database will need to be created beforehand

2. Backend Service (Python FastAPI)

# Navigate to the backend directory
cd backend/app

# Create and activate virtual environment
python -m venv venv
source venv/bin/activate  # On Windows, use: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Set up environment variables in .env file
# Example:
AWS_ACCESS_KEY_ID=your_access_key
AWS_SECRET_ACCESS_KEY=your_secret_key
AWS_REGION=your_aws_region
MODEL_ID=anthropic.claude-3-5-sonnet-20240620-v1:0
DB_USERNAME=your_db_user
DB_PASSWORD=your_db_password
DB_NAME=health_anxiety

# Start the FastAPI server
fastapi dev

3. Frontend (React)

# Navigate to the frontend directory
cd frontend

# Install dependencies
npm install

# Start the development server
npm run start

4. Docker Set-up

# Create a .env file in the root folder with folowing variables
#Example:

# Database Configuration
DB_NAME=your_db_name
DB_USERNAME=your_db_username
DB_PASSWORD=your_db_password
DB_HOST=postgres
DB_PORT=5432

# Superuser Configuration
DB_SUPERUSER_USERNAME=your_superuser_name
DB_SUPERUSER_PASSWORD=your_superuser_password
DB_SUPERUSER_EMAIL=your_superuser_email

# AWS Configuration
AWS_ACCESS_KEY_ID=your_aws_acces_key
AWS_SECRET_ACCESS_KEY=your_aws_secret_acces
AWS_REGION=your_aws_region

# Bedrock Configuration
MODEL_ID=your_model_of_choice
MODEL_PROVIDER=bedrock_converse
MAX_TOKENS=1000
TEMPERATURE=0.3
TOP_P=0.4

# API Configuration
API_HOST=0.0.0.0
API_PORT=8000
LOG_LEVEL=INFO
API_PREFIX=/api/v1

# App settings
APP_NAME=HypochondriAI - Health anxiety specialist
APP_VERSION=0.1.0
APP_DESCRIPTION=AI Agent offering personalized support for people struggling with health anxiety

# The application can now be started locally with a single command
docker compose up --build

License

This project is licensed under the GPL 3.0 License - see the LICENSE file for details.