🧊 Ice Breaker

An intelligent ice breaker generator powered by LangChain and social media intelligence

🎯 Overview

Ice Breaker is a sophisticated AI-powered web application that creates personalized ice breakers by analyzing LinkedIn and Twitter profiles. This project serves as a comprehensive learning tool for mastering LangChain while building a practical generative AI application that combines social media intelligence with natural language generation.

✨ Key Features

AI Pipeline Flow:

1. 🔍 Profile Discovery: Intelligent lookup and discovery of LinkedIn and Twitter profiles
2. 🌐 Data Extraction: Advanced web scraping of professional and social media data
3. 🧠 AI Analysis: Deep analysis of personality, interests, and professional background
4. ✍️ Ice Breaker Generation: Context-aware creation of personalized conversation starters
5. 🎨 Smart Formatting: Professional presentation of generated content
6. 💬 Interactive Interface: User-friendly web interface powered by Flask
7. 🚀 Real-time Processing: Fast end-to-end pipeline from profile input to ice breaker output

Watch Ice Breaker analyze social profiles and generate personalized conversation starters

🛠️ Tech Stack

Component	Technology	Description
🖥️ Frontend	Flask	Web application framework
🧠 AI Framework	LangChain 🦜🔗	Orchestrates the AI pipeline
🔍 LinkedIn Data	Scrapin.io	Professional profile scraping
🐦 Twitter Data	Twitter API	Social media content analysis
🌐 Web Search	Tavily	Enhanced profile discovery
🤖 LLM	OpenAI GPT	Powers the conversation generation
📊 Monitoring	LangSmith	Optional tracing and debugging
🐍 Backend	Python 3.8+	Core application logic

🚀 Quick Start

Prerequisites

• Python 3.8 or higher
• OpenAI API key
• Scrapin.io API key
• Twitter API credentials
• Tavily API key

Installation

1. Clone the repository

   git clone https://github.com/emarco177/ice_breaker.git
   cd ice_breaker

2. Set up environment variables

   Create a .env file in the root directory with your API keys (see Environment Variables section for details).

3. Install dependencies

   pipenv install

4. Start the application

   pipenv run app.py

5. Open your browser and navigate to http://localhost:5000

🧪 Testing

Run the test suite to ensure everything is working correctly:

pipenv run pytest .

💰 API Costs & Credits

📋 Note: This project uses paid API services for optimal functionality:

• Scrapin.io 💼 - LinkedIn data scraping
  Sign up for API access

• Tavily 🌐 - Enhanced web search and profile discovery
  Sign up for Tavily API access

• Twitter API 🐦 - Social media content access
  Paid service for accessing Twitter data

⚠️ Important: If you enable LangSmith tracing (LANGCHAIN_TRACING_V2=true), ensure you have a valid LANGCHAIN_API_KEY. Without it, the application will throw an error. If you don't need tracing, simply omit these variables.

📁 Project Structure

ice_breaker/
├── agents/                    # AI agents for profile lookup
│   ├── linkedin_lookup_agent.py
│   └── twitter_lookup_agent.py
├── chains/                    # LangChain custom chains
│   └── custom_chains.py
├── third_parties/            # External API integrations
│   ├── linkedin.py
│   └── twitter.py
├── tools/                    # Utility tools and functions
│   └── tools.py
├── templates/                # Flask HTML templates
│   └── index.html
├── static/                   # Static assets
│   ├── banner.jpeg
│   └── demo.gif
├── app.py                    # Flask application entry point
├── ice_breaker.py           # Core ice breaker logic
├── output_parsers.py        # Response formatting utilities
└── requirements files       # Pipfile, Pipfile.lock

🔧 Environment Variables

Create a .env file in the root directory:

OPENAI_API_KEY=your_openai_api_key_here
SCRAPIN_API_KEY=your_scrapin_api_key_here
TAVILY_API_KEY=your_tavily_api_key_here

# Optional: Twitter scraping (if you want Twitter data)
TWITTER_API_KEY=your_twitter_api_key_here
TWITTER_API_SECRET=your_twitter_api_secret_here
TWITTER_ACCESS_TOKEN=your_twitter_access_token_here
TWITTER_ACCESS_SECRET=your_twitter_access_secret_here

# Optional: Enable LangSmith tracing
LANGCHAIN_TRACING_V2=true
LANGCHAIN_API_KEY=your_langsmith_api_key_here
LANGCHAIN_PROJECT=ice_breaker

⚠️ Important Note: If you enable tracing by setting LANGCHAIN_TRACING_V2=true, you must have a valid LangSmith API key set in LANGCHAIN_API_KEY. Without a valid API key, the application will throw an error. If you don't need tracing, simply remove or comment out these environment variables.

Variable	Description	Required
OPENAI_API_KEY	Your OpenAI API key for LLM access	✅
SCRAPIN_API_KEY	Scrapin.io API key for LinkedIn scraping	✅
TAVILY_API_KEY	Tavily API key for enhanced web search	✅
TWITTER_API_KEY	Twitter API key for social data access (optional)	⚪
TWITTER_API_SECRET	Twitter API secret (optional)	⚪
TWITTER_ACCESS_TOKEN	Twitter access token (optional)	⚪
TWITTER_ACCESS_SECRET	Twitter access token secret (optional)	⚪
LANGCHAIN_TRACING_V2	Enable LangSmith tracing (optional)	⚪
LANGCHAIN_API_KEY	LangSmith API key (required if tracing enabled)	⚪
LANGCHAIN_PROJECT	LangSmith project name (optional)	⚪

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

📚 Learning Resources

This project is designed as a comprehensive learning tool for understanding:

• 🦜 LangChain Framework - Agent orchestration and chain composition
• 🔗 API Integration - Working with multiple external services
• 🧠 AI Application Architecture - Building production-ready AI systems
• 🌐 Web Scraping - Ethical data collection from social platforms
• 💬 Natural Language Generation - Context-aware content creation

📄 License

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

🌟 Support

If you find this project helpful, please consider:

• ⭐ Starring the repository
• 🐛 Reporting issues
• 💡 Contributing improvements
• 📢 Sharing with others
• 🎓 Taking the LangChain Course

---

🔗 Connect with Me

Built with ❤️ by Eden Marco