An empathetic AI companion system designed specifically for elderly care, featuring nurse-like persona, memory retention, health monitoring, and voice-first interface.
- ✅ User Authentication - Register, login, logout with JWT tokens
- ✅ AI Chat - Conversation with Claude 3.5 Sonnet (nurse persona)
- ✅ Mobile App - Complete UI with login, chat, and session management
- ✅ Conversation History - View and manage past conversations
- ✅ Auto Token Refresh - Seamless authentication experience
- ✅ Test User - Ready-to-use test account (see TEST_USER.md)
- ✅ API Documentation - Interactive Swagger UI at
/docs - ✅ Database - PostgreSQL with 12 tables initialized
Try it now: Clone the repo, start Docker services, run the backend, and launch the mobile app!
Smart AI provides elderly individuals with a compassionate, intelligent companion that enhances their quality of life through:
- 🗣️ Voice-First Interface - Natural conversation via speech
- 🧠 Memory Retention - Remembers past conversations and context
- 💊 Health Monitoring - Tracks health signals from conversations
- 👨⚕️ Caregiver Integration - Alerts and insights for family/caregivers
- 📱 Offline-First - Works without constant internet connection
- 🔒 HIPAA Compliant - Enterprise-grade security and privacy
seva-ai/
├── docs/ # Comprehensive documentation
├── database/ # Database schemas and migrations
├── backend/ # FastAPI backend service
├── mobile-app/ # React Native mobile application
├── admin-dashboard/ # Caregiver web dashboard
├── shared/ # Shared code and types
├── infrastructure/ # DevOps and deployment configs
├── .github/workflows/ # CI/CD pipelines
├── PLAN.md # Project roadmap
└── TASKS.md # Task breakdown
| Directory | Description | Tech Stack | Status |
|---|---|---|---|
| docs/ | Architecture, API docs, guides | Markdown | ✅ Complete |
| database/ | PostgreSQL & Weaviate schemas | SQL, Python | ✅ Schema Complete |
| backend/ | API service, Claude AI integration | FastAPI, Python 3.11+ | ✅ Phase 1 Complete |
| mobile-app/ | iOS/Android app for elderly users | React Native, Expo, TypeScript | ✅ Phase 1 Complete |
| admin-dashboard/ | Caregiver monitoring portal | React, Vite, Tailwind CSS | 🚧 Scaffold Ready |
| infrastructure/ | Docker Compose, CI/CD | Docker, GitHub Actions | ✅ Dev Environment Ready |
- Node.js 18+ (for mobile app and dashboard)
- Python 3.11+ (for backend)
- Docker & Docker Compose (for local development)
- PostgreSQL 15+ (or use Docker)
- Weaviate (or use Docker)
-
Clone the repository
git clone https://github.com/MindweaveTech/seva-ai.git cd seva-ai -
Start infrastructure services
docker-compose -f infrastructure/docker/docker-compose.dev.yml up -d # Starts PostgreSQL, Weaviate, Redis, Adminer -
Setup backend
cd backend python -m venv venv source venv/bin/activate # Windows: venv\Scripts\activate pip install -r requirements.txt cp .env.example .env # Add your ANTHROPIC_API_KEY to .env # Database is already initialized! # Or run: alembic upgrade head uvicorn app.main:app --reload # Backend runs at http://localhost:8000 # API docs at http://localhost:8000/docs
-
Setup mobile app
cd mobile-app npm install # Backend URL is pre-configured for localhost npm start # Start Expo # Press 'i' for iOS or 'a' for Android
-
Login with test user
Email: test@smartai.com Password: TestPass123!See TEST_USER.md for details.
-
Setup admin dashboard
cd admin-dashboard npm install cp .env.example .env npm run dev # Start dev server
- Architecture Guide - Complete system architecture
- Implementation Guide - Step-by-step development guide
- Database Setup - Database schemas and migrations
- Technical Summary - Quick reference guide
- Project Plan - Roadmap, timeline, team structure
- Task Breakdown - Detailed task list (300+ tasks)
- React Native with Expo ~49.0 - Cross-platform mobile development
- TypeScript 5.1+ (strict mode) - Type safety
- Zustand 4.4+ - Lightweight state management
- React Navigation 6.1+ - Stack navigation
- Axios 1.6+ - HTTP client with interceptors
- AsyncStorage - Persistent token storage
- FastAPI - Modern async Python web framework
- SQLAlchemy 2.0+ - ORM with async support
- Pydantic v2 - Data validation and settings
- Alembic - Database migrations
- Anthropic Python SDK - Claude API integration
- Uvicorn - ASGI server
- PostgreSQL 15+ - Primary relational database (12 tables)
- Weaviate 1.22.4 - Vector database for RAG (Phase 2)
- Redis 7 - Caching and sessions (optional)
- Claude 3.5 Sonnet (2024-10-22) - Latest Anthropic conversational AI
- OpenAI Embeddings - Text embeddings (Phase 2)
- Expo AV - Audio recording/playback (Phase 3)
- Kubernetes - Container orchestration
- Docker - Containerization
- GitHub Actions - CI/CD
- Prometheus & Grafana - Monitoring
- 🎙️ Voice Conversation - Speak naturally, get voice responses
- 💭 Memory - AI remembers your health history and preferences
- 🏥 Health Tracking - Gentle monitoring through conversation
- 📵 Offline Mode - Works without internet connection
- 🌙 Daily Check-ins - Regular wellness conversations
- 🚨 Emergency Support - Quick access to caregivers
- 📊 Health Dashboard - Monitor patient wellness
- 🔔 Smart Alerts - Get notified of concerning patterns
- 📈 Health Trends - Visualize health metrics over time
- 💬 Conversation Summaries - Review key discussion points
- 👥 Multi-Patient Management - Monitor multiple elderly users
- 📱 Mobile & Web Access - Check in from anywhere
- Project setup and architecture
- Development environment (Docker Compose)
- Database initialization (12 tables)
- CI/CD pipelines (GitHub Actions)
- Backend/Mobile/Dashboard scaffolding
- JWT authentication (access + refresh tokens)
- User registration and login
- Claude 3.5 Sonnet integration
- Chat API with conversation history
- Mobile UI (Login, Register, Chat, Sessions)
- Navigation and state management
- Auto token refresh
- Test user created
- Weaviate vector database integration
- Conversation embeddings
- Semantic search
- Long-term memory retrieval
- Speech-to-text
- Text-to-speech
- Audio recording
- Voice commands
- WatermelonDB integration
- Sync engine
- Offline message queue
- Conflict resolution
- Health metrics tracking
- Sentiment analysis
- Pattern detection
- Caregiver alerts
- Dashboard UI implementation
- User management
- Health monitoring visualization
- Analytics and reporting
- Comprehensive testing
- Security hardening
- HIPAA compliance audit
- Production deployment
See PLAN.md for detailed roadmap.
We welcome contributions! Please see our Contributing Guide for details.
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
- Backend: Python 3.11+, Black formatting, 90%+ test coverage
- Frontend: TypeScript strict mode, ESLint, 85%+ test coverage
- Commits: Conventional Commits format
- PRs: Require 1+ approval, all tests passing
- Backend Engineers - API development, database, RAG pipeline
- Mobile Engineers - React Native app, offline sync
- Frontend Engineer - Admin dashboard
- ML Engineer - RAG optimization, health analytics
- DevOps Engineer - Infrastructure, CI/CD, monitoring
- QA Engineer - Testing, quality assurance
See PLAN.md for detailed team structure.
This project is licensed under the MIT License - see the LICENSE file for details.
Security is paramount when handling health data. We follow industry best practices:
- 🔐 Encryption at rest (AES-256)
- 🔒 Encryption in transit (TLS 1.3)
- 🎫 JWT authentication with refresh tokens
- 📝 Audit logging for all sensitive operations
- 🛡️ HIPAA compliance framework
- 🔍 Regular security audits
Report security vulnerabilities: security@mindweavetech.com (or use GitHub Security Advisories)
- Documentation: docs/
- Issues: GitHub Issues
- Discussions: GitHub Discussions
- Email: support@mindweavetech.com
- Anthropic - Claude API for empathetic AI conversations
- OpenAI - Embedding models for semantic search
- Google Cloud - Speech services
- React Native Community - Mobile framework and ecosystem
- FastAPI Community - Modern Python web framework
Current Phase: Phase 0 - Foundation & Architecture ✅ Next Milestone: Development environment setup Timeline: 20-week MVP development Last Updated: 2025-10-20
Built with ❤️ by MindweaveTech
Making elderly care more compassionate through AI