An intelligent life coach & productivity strategist powered by multi-agent AI architecture
Download App (APK) • Live API • Report Bug • Request Feature
- Overview
- Key Features
- Architecture
- Tech Stack
- Getting Started
- Project Structure
- API Documentation
- Mobile Application
- Multi-Agent System
- Database Schema
- Deployment
- Development
- Design Principles
- Contributing
- Roadmap
- License
- Acknowledgments
NEEL transcends traditional productivity tracking by functioning as a sophisticated AI life coach that monitors work patterns, calibrates strategy, and delivers high-level insights through a premium glassmorphic interface.
Designed for high-performers — Engineers, Researchers, Creators — NEEL serves as a cognitive partner to optimize deep-work patterns and long-term trajectory through intelligent behavioral analysis and personalized coaching.
- 🧠 Multi-Agent Intelligence: Supervisor, Reasoning, and Reflection agents work in harmony
- 💾 Persistent Memory: Contextual awareness across all conversations
- 🪄 Magic Auto-Logging: Natural language activity parsing
- 📊 Behavioral Analytics: Deep insights into productivity patterns
- 🎨 Premium UX: Glassmorphic design with dynamic gradients
- ☁️ Cloud-Native: Production-ready deployment on Vercel + Supabase
A real-time command center displaying your Calibration Status with dynamic activity breakdowns revealing exactly where cognitive energy is being spent.
NEEL remembers your evolution. Chat history persists in the cloud, enabling the AI to reference past goals and wins for context-aware coaching.
Log work without forms. Simply tell NEEL what you accomplished, and AI automatically parses, categorizes, and logs the activity.
Multi-agent reasoning system identifies behavioral patterns, suggests optimizations, and provides strategic guidance based on your historical data.
Track progress toward goals with intelligent calibration that adapts recommendations based on your evolving productivity patterns.
All data encrypted in transit and at rest. User data isolation with secure authentication.
NEEL implements a backend-first, multi-agent AI architecture with cloud-native deployment:
┌─────────────────────────────────────────────────────────┐
│ Mobile Client (React Native) │
│ Glassmorphic UI • Expo SDK 54 │
└────────────────────┬────────────────────────────────────┘
│ REST API
▼
┌─────────────────────────────────────────────────────────┐
│ FastAPI Backend (Python 3.11+) │
│ ┌─────────────────────────────────────────────────┐ │
│ │ Multi-Agent Orchestration Layer │ │
│ │ ┌──────────┐ ┌──────────┐ ┌──────────────┐ │ │
│ │ │Supervisor│ │ Reasoning│ │ Reflection │ │ │
│ │ │ Agent │──│ Agent │──│ Agent │ │ │
│ │ └──────────┘ └──────────┘ └──────────────┘ │ │
│ └─────────────────────────────────────────────────┘ │
│ │ │
│ ┌──────────┴──────────┐ │
│ ▼ ▼ │
│ ┌─────────────────┐ ┌──────────────┐ │
│ │ LangChain Core │ │ Gemini 1.5 │ │
│ │ Orchestration │ │ Flash │ │
│ └─────────────────┘ └──────────────┘ │
└────────────────────┬────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────┐
│ PostgreSQL Database (Supabase Managed) │
│ Users • Activities • Chats • Analytics │
└─────────────────────────────────────────────────────────┘
| Agent | Role | Responsibility |
|---|---|---|
| Supervisor | Gatekeeper | Validates data sufficiency before reasoning |
| Reasoning | Brain | Identifies trends, generates insights & guidance |
| Reflection | Auditor | Ensures safety, tone quality, and appropriateness |
- Framework: FastAPI 0.104+
- AI/ML: LangChain + Google Gemini 1.5 Flash
- Database: PostgreSQL 15+ (Alembic migrations)
- ORM: SQLAlchemy 2.0+
- Deployment: Vercel (Serverless Cloud Platform)
- Authentication: JWT-based secure auth
- API Docs: Auto-generated OpenAPI (Swagger)
- Framework: React Native (Expo SDK 54)
- Language: TypeScript + JavaScript
- UI Libraries: Lucide React Native Icons
- Build: EAS Build (Production APK)
- State Management: React Hooks + Context
- Styling: Glassmorphic design system
- Version Control: Git + GitHub
- Migrations: Alembic
- Process Manager: Uvicorn + Gunicorn
- Testing: Advanced diagnostic scripts
- Notebooks: Jupyter for AI experimentation
- Python 3.11+
- Node.js 18+
- PostgreSQL 15+
- Google Gemini API Key
- Expo CLI (for mobile development)
- Clone the repository
git clone https://github.com/Edge-Explorer/NEEL.git
cd NEEL- Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate- Install dependencies
pip install -r requirements.txt- Configure environment variables
cp .env.example .env
# Edit .env and add your credentials:
# - DATABASE_URL
# - GOOGLE_GEMINI_API_KEY
# - SECRET_KEY- Run database migrations
alembic upgrade head- Seed initial data (optional)
python seed_data.py- Start the development server
uvicorn backend.main:app --host 0.0.0.0 --port 8000 --reloadAPI will be available at: http://localhost:8000/docs
- Navigate to mobile directory
cd mobile- Install dependencies
npm install- Configure API endpoint
// mobile/services/api.js
const API_BASE_URL = 'http://YOUR_LOCAL_IP:8000'; // For local dev
// OR
const API_BASE_URL = 'https://neel-green.vercel.app'; // For production- Start Expo development server
npx expo start- Run on device
- Scan QR code with Expo Go app (Android/iOS)
- Or press
afor Android emulator - Or press
ifor iOS simulator
cd mobile
eas build --platform android --profile productionNEEL/
├── backend/ # FastAPI backend application
│ ├── main.py # Application entry point
│ ├── models/ # SQLAlchemy database models
│ ├── routers/ # API route handlers
│ ├── services/ # Business logic layer
│ ├── agents/ # Multi-agent AI system
│ └── utils/ # Utility functions
│
├── mobile/ # React Native mobile app
│ ├── App.js # Mobile app entry point
│ ├── components/ # Reusable UI components
│ ├── screens/ # Application screens
│ ├── services/ # API integration layer
│ ├── assets/ # Images, fonts, icons
│ └── app.json # Expo configuration
│
├── alembic/ # Database migration scripts
│ ├── versions/ # Migration version files
│ └── env.py # Alembic environment config
│
├── notebooks/ # Jupyter notebooks for AI experiments
├── scripts/ # Utility scripts
├── models/ # Serialized AI models
├── data/ # Data files and datasets
├── ai_design/ # AI-generated design assets
│
├── requirements.txt # Python dependencies
├── alembic.ini # Alembic configuration
├── render.yaml # Render deployment config
├── Procfile # Process configuration
├── start.sh # Startup script
└── README.md # This file
- Production:
https://neel-green.vercel.app - Local:
http://localhost:8000
- Swagger UI:
/docs - ReDoc:
/redoc
POST /api/auth/register
POST /api/auth/login
POST /api/auth/refreshGET /api/users/me
PUT /api/users/me
DELETE /api/users/mePOST /api/activities/ # Create activity
GET /api/activities/ # List activities
GET /api/activities/{id} # Get activity detail
PUT /api/activities/{id} # Update activity (24hr window)
DELETE /api/activities/{id} # Delete activityPOST /api/chat/message # Send message to AI
GET /api/chat/history # Retrieve chat history
DELETE /api/chat/clear # Clear chat historyGET /api/analytics/dashboard # Dashboard metrics
GET /api/analytics/trends # Productivity trends
GET /api/analytics/calibration # Calibration statuscurl -X POST "https://neel-green.vercel.app/api/activities/auto-log" \
-H "Authorization: Bearer YOUR_JWT_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"description": "Spent 3 hours implementing the authentication system and writing unit tests"
}'Response:
{
"id": "uuid-here",
"category": "Development",
"duration": 180,
"parsed_description": "Implemented authentication system and wrote unit tests",
"timestamp": "2024-01-22T10:30:00Z"
}NEEL's mobile interface follows a glassmorphic design system with:
- 🌑 Dark mode by default
- 🔮 Glass-like translucent components
- 🎨 Dynamic linear gradients
- ✨ Smooth animations and transitions
- 📐 Consistent spacing and typography
- Dashboard - Real-time productivity pulse
- Activity Logger - Manual and auto-logging
- Chat Interface - AI coaching conversations
- Analytics - Visual insights and trends
- Profile - User settings and preferences
{
"expo": {
"name": "NEEL",
"slug": "neel",
"version": "1.0.0",
"sdkVersion": "54.0.0",
"platforms": ["ios", "android"],
"android": {
"package": "com.karan.neel",
"adaptiveIcon": {
"foregroundImage": "./assets/images/adaptive-icon.png"
}
}
}
}NEEL's intelligence is powered by a three-tier agent system:
# Simplified Agent Flow
class SupervisorAgent:
"""Gates reasoning based on data sufficiency"""
def evaluate(self, user_data):
if len(user_data.activities) < MIN_THRESHOLD:
return "insufficient_data"
return "proceed_to_reasoning"
class ReasoningAgent:
"""Core intelligence for trend analysis"""
def analyze(self, user_data):
patterns = self.identify_patterns(user_data)
insights = self.generate_insights(patterns)
recommendations = self.create_recommendations(insights)
return recommendations
class ReflectionAgent:
"""Quality assurance and safety auditor"""
def audit(self, ai_response):
safety_check = self.verify_safety(ai_response)
tone_check = self.verify_tone(ai_response)
if not (safety_check and tone_check):
return self.modify_response(ai_response)
return ai_responseThe agents communicate through LangChain's graph-based execution:
- User message → Supervisor Agent (data validation)
- If approved → Reasoning Agent (insight generation)
- Output → Reflection Agent (quality audit)
- Final response → User
-- Users table
CREATE TABLE users (
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
email VARCHAR(255) UNIQUE NOT NULL,
hashed_password VARCHAR(255) NOT NULL,
full_name VARCHAR(255),
created_at TIMESTAMP DEFAULT NOW(),
updated_at TIMESTAMP DEFAULT NOW()
);
-- Activities table
CREATE TABLE activities (
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
user_id UUID REFERENCES users(id) ON DELETE CASCADE,
category VARCHAR(100) NOT NULL,
description TEXT,
duration INTEGER, -- minutes
logged_at TIMESTAMP DEFAULT NOW(),
created_at TIMESTAMP DEFAULT NOW(),
can_edit_until TIMESTAMP -- 24-hour edit window
);
-- Chat history
CREATE TABLE chat_messages (
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
user_id UUID REFERENCES users(id) ON DELETE CASCADE,
role VARCHAR(20) NOT NULL, -- 'user' or 'assistant'
content TEXT NOT NULL,
created_at TIMESTAMP DEFAULT NOW()
);
-- Analytics snapshots
CREATE TABLE analytics_snapshots (
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
user_id UUID REFERENCES users(id) ON DELETE CASCADE,
calibration_status VARCHAR(50),
total_activities INTEGER,
most_productive_category VARCHAR(100),
snapshot_date DATE,
created_at TIMESTAMP DEFAULT NOW()
);# Create new migration
alembic revision --autogenerate -m "description"
# Apply migrations
alembic upgrade head
# Rollback
alembic downgrade -1NEEL is deployed on Vercel as a Serverless FastAPI application.
vercel.json
{
"functions": {
"api/main.py": {
"maxDuration": 10
}
},
"rewrites": [
{
"source": "/(.*)",
"destination": "api/main.py"
}
]
}The production database is hosted on Supabase for its excellent performance and connection pooling (essential for serverless functions).
Required environment variables for production:
DATABASE_URL=postgresql://user:pass@host:5432/dbname
GOOGLE_GEMINI_API_KEY=your_gemini_api_key
SECRET_KEY=your_secret_key_for_jwt
ALGORITHM=HS256
ACCESS_TOKEN_EXPIRE_MINUTES=30- Connect GitHub repository to Render
- Configure environment variables
- Deploy backend service
- Run migrations (auto-run via start script)
- Verify deployment at
/docsendpoint
curl https://neel-green.vercel.app/docsExpected response:
{
"status": "healthy",
"database": "connected",
"ai_service": "operational"
}- Create feature branch
git checkout -b feature/amazing-feature- Make changes and test
# Backend tests
python -m pytest tests/
# Frontend lint
cd mobile && npm run lint- Commit with conventional commits
git commit -m "feat: add amazing feature"- Push and create PR
git push origin feature/amazing-featureNEEL maintains high code quality through:
- Type hints in Python (mypy)
- Linting: flake8, black, isort
- Testing: pytest with >80% coverage
- Documentation: Docstrings for all public APIs
# Run all tests
pytest
# Run with coverage
pytest --cov=backend --cov-report=html
# Run specific test file
pytest tests/test_agents.py- Aesthetic First - Every UI change follows the glassmorphic design system
- 24-Hour Rule - Activity logs adjustable within 24 hours for data integrity
- Agent Audit - Every AI response audited before delivery
- Privacy by Design - User data encrypted and isolated
- Performance Matters - <100ms API response time target
- Fail Gracefully - Clear error messages and recovery paths
- Use glassmorphic cards for all content containers
- Maintain 16px base spacing unit
- Linear gradients for emphasis (purple → blue)
- Smooth transitions (200-300ms cubic-bezier)
- Dark mode color palette: #0A0E27 base
- Minimum 4.5:1 contrast ratio for accessibility
Contributions make the open-source community an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/AmazingFeature) - Commit your Changes (
git commit -m 'Add some AmazingFeature') - Push to the Branch (
git push origin feature/AmazingFeature) - Open a Pull Request
- Follow the existing code style
- Write meaningful commit messages
- Add tests for new features
- Update documentation
- Ensure all tests pass before submitting PR
- 🐛 Bug fixes and issue resolution
- ✨ New AI agent capabilities
- 📱 Mobile UI enhancements
- 📊 New analytics visualizations
- 🌍 Internationalization (i18n)
- 📝 Documentation improvements
- 🧪 Test coverage expansion
- iOS app support
- Voice input for activity logging
- Weekly productivity reports
- Team collaboration features
- Advanced pattern recognition
- Integration with calendar apps
- Custom goal templates
- Export analytics to CSV/PDF
- Multi-model AI support (Claude, GPT-4)
- Predictive scheduling assistant
- Habit formation tracking
- Social accountability features
See the open issues for a full list of proposed features and known issues.
Distributed under the MIT License. See LICENSE for more information.
MIT License
Copyright (c) 2024 Karan (Edge-Explorer)
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software...
- FastAPI - Modern Python web framework
- LangChain - AI orchestration framework
- Google Gemini - Large language model
- React Native - Mobile framework
- Expo - React Native toolchain
- Vercel - Cloud platform
- Supabase - Database system
Developer: Karan (Edge-Explorer)
- GitHub: @Edge-Explorer
- Project Link: https://github.com/Edge-Explorer/NEEL
- Issues: Report a bug