Family Med Nanny Service Architecture

TODO: Description of the application...

System Overview

This document describes the architecture of the Family Med Nanny service, a comprehensive platform for managing family medical needs.

Keep this link handy: Icons for mermaid architecture diagrams

High-Level Architecture

This hides a single Mermaid code block that github has issues rendering correctly. The above PNG is the correct representation of the diagram as it should render. Including a PNG of the diagram is an efficient solution. If you want to see the error, take a look.

---
config:
  theme: dark
  # theme: neutral
---
architecture-beta
    service slack(logos:slack-icon)[Slack MedNannyAI]
    service whatsapp(logos:whatsapp-icon)[WhatsApp MedNannyAI]
    service twilio(logos:twilio-icon)[Twilio]
    service fastapi(logos:fastapi-icon)[API]
    service core(logos:python)[Core]
    service auth(logos:auth0-icon)[Authn Authz]
    service ai("<img src='https://avatars.githubusercontent.com/u/110818415' style='background-color:black;vertical-align:middle;margin:0px 0px'>")[MedNannyAI Assistant]
    service llm(logos:anthropic-icon)[LLM Provider]
    service db(logos:sqlite)

    junction frontendcenter
    junction frontendleft
    junction frontendright

    whatsapp:L <-- R:twilio
    frontendleft:T -- B:slack
    frontendright:T -- B:twilio
    frontendleft:R -- L:frontendcenter
    frontendcenter:R -- L:frontendright
    fastapi:B --> T:core
    fastapi:R --> L:auth
    core:L --> R:ai
    ai:L --> R:llm
    frontendcenter:B -- T:fastapi
    core:B --> T:db

    %%group frontend(cloud)[MedNannyAI Frontend]
    %%group twilio_whatsapp(logos:whatsapp-icon)[WhatsApp MedNannyAI] in frontend
    %%group backend(logos:fastapi-icon)[MedNannyAI API Backend]
    %%frontendCenter:B -- T:fastapi{group}
    %%slack{group}:B -- T:fastapi{group}
    %%whatsapp{group}:B -- T:fastapi{group}
    %%group data(logos:aws-lambda)[Data Persistance]

Loading

Service Components

Frontend Layer

• Slack Integration: Real-time messaging interface on mobile/desktop/web
• WhatsApp Integration: Mobile messaging interface
• Twilio Integration: SMS messages using webhooks

Backend Services

• Authentication Service: User authentication and authorization
• User Management: User profiles and account management
• Medical Records: Secure storage and management of medical information
• AI Assistant: Intelligent assistance for medical queries and recommendations

Data Layer

• Database: sqlite for structured data storage
• Cache: Redis (maybe?) for session management and performance optimization

Data Flow

---
config:
  theme: dark
  # theme: neutral
---
sequenceDiagram
    participant U as User
    participant F as Frontend
    participant S as Backend Service
    participant D as Database
    participant E as External Service

    U->>F: User Action
    F->>S: API Request
    S->>S: Authenticate & Authorize
    S->>D: Query/Update Data
    D-->>S: Response
    S->>E: External Service Call (if needed)
    E-->>S: External Response
    S-->>F: API Response
    F-->>U: Update UI

Loading

Security Architecture

---
config:
  theme: dark
  # theme: neutral
---
flowchart TB
    subgraph "Data Protection"
        direction TB
        PII(PII Protection)-->HIPAA(HIPAA Compliance)
        HIPAA-->GDPR(GDPR Compliance)
    end

    subgraph "Security Layers"
        direction TB
        Auth(Authentication)-->Authz(Authorization)
        Authz-->Encrypt(Encryption)
        Encrypt-->Audit(Audit Logging)
    end

Loading

Deployment Architecture

---
config:
  layout: elk
  elk:
    mergeEdges: false
    nodePlacementStrategy: SIMPLE
  theme: dark
  # theme: neutral
---
flowchart TB
    subgraph "Application Layer"
        Application
    end

    subgraph "Data Layer"
        DB[(Database)]
        Cache[(Cache)]
    end

    Application --> DB
    Application --> Cache

Loading

Technology Stack

Frontend

• Slack API Integration
• WhatsApp Business API

Backend

• FastAPI
• sqlite
• Pydantic AI
• Anthropic and/or OpenAI
• Redis (maybe, probably in memory caching or just go with sql db caching...it's fast enough)

Infrastructure

• Docker
• Railway for deployment and monitoring
• Twilio for SMS Integration uses webhooks to handle incoming/outgoing SMS messages
• Single server deployment

Monitoring

• Railway should give me what I need for something of this scope out of the box.

Next Steps

This architecture will be iteratively refined based on:

1. Specific requirements gathering
2. Performance requirements
3. Security requirements
4. Integration requirements

Future Enhancements

• Database integration (SQLAlchemy)
• User authentication
• Medication reminders
• Drug interaction warnings
• Family member access
• Export functionality
• Photo logging
• Integration with health apps

License

MIT License - feel free to use this for your medication management needs!