LinkGenie is a full-stack web application that leverages the power of Large Language Models (LLMs) to streamline LinkedIn outreach. It allows users to search for LinkedIn profiles by name and retrieve relevant information. LinkGenie then utilizes LLMs to personalize a "connection link" with the retrieved information, facilitating conversation starters. This allows users to initiate meaningful conversations with their desired connections.

1. LinkedIn Data Scraping

• Proxycurl API Integration: Efficiently scrapes LinkedIn data about a person, gathering valuable information for networking purposes.

2. Advanced Search Capabilities

• Tavily API: Employed as a search API, providing high-quality, real-time knowledge optimized for RAG and LLMs.

3. LangChain Integration

• Chains and Agents: Utilizes LangChain's chains and agents, including custom agents, to efficiently handle and process data.
• Tools and Custom Tools: Implements both standard and custom tools within the application to enhance its functionality.
• Output Parsers: Uses output parsers to format and display the retrieved information in a user-friendly manner.

4. Flask API Integration

• Backend Development: Employs Flask API to build the backend of the application, ensuring smooth and efficient data handling and processing.

To run this project, you will need to add the following environment variables to your .env file

OPENAI_API_KEY

PROXYCURL_API_KEY

TAVILY_API_KEY

LANGCHAIN_API_KEY

LANGCHAIN_TRACING_V2=true

LANGCHAIN_PROJECT=Link Genie

1. Clone the repository to your local environnement

git clone https://github.com/chaeyeon2367/llm-app-LinkGenie.git

2. Go to the project directory

cd llm-app-LinkGenie

3. Install dependencies

pipenv install

4. Start the flask server

pipenv run app.py

To run tests, run the following command

pipenv run pytest .

https://github.com/hwchase17/chat-langchain-readthedocs

https://github.com/emarco177/ice_breaker