Agentic AI News Platform

  1. ➜ GitHub Repository Link

Tech News Digest

A Personalized AI-Powered News Aggregation.

A sophisticated tech news aggregation platform that delivers personalized daily and weekly digests using AI-powered content categorization and email notifications.

Features

Personalized Digests & Notifications

  • Daily Digests: Personalized morning summaries of tech news
  • Weekly Roundups: Comprehensive weekly reports with trending topics
  • Instant Notifications: Breaking news alerts for high-impact stories
  • Smart Categorization: AI-powered content classification into 10 tech categories
  • Custom Keywords: Subscribe to specific topics with custom keyword filters

AI-Powered Intelligence

  • Content Categorization: Automatic article classification using keyword matching and relevance scoring
  • Smart Summaries: AI-generated article summaries using Together AI
  • Personalization Engine: Content recommendation based on user interests
  • Interactive Chat: Ask questions about recent tech news with AI assistant

User Management

  • User Authentication: JWT-based secure authentication
  • Preference Management: Customizable digest frequency and topics
  • Subscription Control: Granular control over notification types
  • Usage Analytics: Track digest history and engagement

Architecture

├── webapp.py              # Main FastAPI application with web UI and API
├── models.py              # SQLAlchemy database models
├── database.py            # Database configuration and initialization
├── auth.py                # JWT authentication and user management
├── schemas.py             # Pydantic models for API validation
├── scheduler_service.py   # APScheduler for automated digest sending
├── services/
│   ├── categorization_service.py  # AI content categorization
│   ├── digest_service.py          # Digest generation and sending
│   └── email_service.py           # Email templates and SMTP
├── scrapers/
│   └── techcrunch.py      # News scraping and processing
└── data/summaries/        # Processed article storage

Quick Start

1. Installation

# Clone the repository
git clone https://github.com/Vishakha-Sawra/agentic-ai-news-platform
cd tech-news-digest

# Install dependencies
pip install -r requirements.txt

2. Environment Setup

# Copy example environment file
cp .env.example .env

# Edit .env with your configuration
nano .env

Required environment variables:

  • TOGETHER_API_KEY: For AI summaries and chat
  • SMTP_*: Email configuration for digest delivery
  • SECRET_KEY: JWT token encryption (generate a secure key)

3. Database Initialization

# Initialize database and create default categories
python database.py

4. Start the Application

# Start the web application
uvicorn webapp:app --reload --host 0.0.0.0 --port 8000

5. Start the Scheduler (Optional)

# Start automated digest scheduler in a separate terminal
python scheduler_service.py

Usage

Web Interface

Visit http://localhost:8000 to access the web interface where you can:

  • Browse categorized tech news articles
  • Subscribe to personalized digests
  • Chat with the AI assistant about recent news

API Documentation

Interactive API documentation is available at:

  • Swagger UI: http://localhost:8000/docs
  • ReDoc: http://localhost:8000/redoc

Key API Endpoints

Authentication
# Register a new user
POST /auth/register
{
  "email": "[email protected]",
  "password": "securepassword",
  "full_name": "John Doe"
}

# Login and get access token
POST /auth/token
username=[email protected]&password=securepassword
Preferences Management
# Get user preferences
GET /preferences
Authorization: Bearer <token>

# Update digest preferences
PUT /preferences
Authorization: Bearer <token>
{
  "daily_digest_enabled": true,
  "weekly_digest_enabled": true,
  "instant_notifications": false,
  "digest_time": "09:00",
  "interested_categories": [1, 2, 3]
}
Digest Operations
# Preview personalized digest
GET /digest/preview?digest_type=daily
Authorization: Bearer <token>

# Send digest immediately
POST /digest/send?digest_type=daily
Authorization: Bearer <token>

# Get digest history
GET /digest/history
Authorization: Bearer <token>
Content Access
# Get all articles
GET /articles?category_id=1&limit=20

# Get personalized articles
GET /articles/personalized?limit=20
Authorization: Bearer <token>

# Get available categories
GET /categories

Content Categories

The system automatically categorizes articles into these domains:

  1. Artificial Intelligence - AI, ML, automation, neural networks
  2. Startups & Funding - Venture capital, funding rounds, IPOs
  3. Big Tech - Google, Apple, Microsoft, Amazon, Meta news
  4. Cybersecurity - Security breaches, privacy, data protection
  5. Mobile & Apps - Smartphones, mobile apps, tablets
  6. Enterprise & SaaS - Business software, productivity tools
  7. Gaming - Video games, esports, gaming platforms
  8. Electric Vehicles - EVs, autonomous driving, transportation
  9. Fintech - Cryptocurrency, digital payments, blockchain
  10. Social Media - Social platforms, creator economy

Configuration

Digest Scheduling

Default schedule (configurable in scheduler_service.py):

  • Daily Digests: 9:00 AM every day
  • Weekly Digests: Monday 9:00 AM
  • Article Sync: Every hour

Email Templates

Professional HTML email templates with:

  • Responsive design for mobile and desktop
  • Category-based organization
  • Personalized content based on user interests
  • Unsubscribe and preference management links

Categorization Algorithm

The AI categorization system:

  1. Extracts keywords from article title, summary, and AI summary
  2. Matches against predefined category keywords
  3. Calculates relevance scores (1-10 scale)
  4. Assigns articles to top 3 matching categories
  5. Only includes categories with relevance score ≥ 3

Advanced Features

Custom Keyword Subscriptions

Users can create custom subscriptions with specific keywords:

POST /subscriptions
Authorization: Bearer <token>
{
  "subscription_type": "daily",
  "keywords": ["quantum computing", "machine learning", "startup funding"]
}

Instant Notifications

High-impact articles (relevance score ≥ 7) trigger instant notifications to subscribed users.

Content Personalization

The system personalizes content by:

  • User-selected category interests
  • Custom keyword subscriptions
  • Historical engagement patterns
  • Relevance scoring algorithms

Benefits

For Users

  • Time-Saving: Curated, personalized content delivery
  • Comprehensive Coverage: Never miss important tech news
  • Smart Filtering: Focus on topics that matter to you
  • Flexible Delivery: Choose your preferred digest frequency

For Organizations

  • Employee Engagement: Keep teams informed about industry trends
  • Market Intelligence: Track competitor and industry developments
  • Knowledge Sharing: Centralized tech news for research teams
  • Customizable: Adapt to specific organizational interests

Production Deployment

Environment Variables

# Production settings
SECRET_KEY=your_production_secret_key
DATABASE_URL=postgresql://user:pass@host:port/dbname
SMTP_SERVER=your_production_smtp_server

Docker Deployment

FROM python:3.9-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY . .
CMD ["uvicorn", "webapp:app", "--host", "0.0.0.0", "--port", "8000"]

Monitoring

  • API endpoint health checks
  • Email delivery monitoring
  • User engagement analytics
  • System performance metrics

Topics