/RateIn

Linkedin profile analyzer - built using openai assistants api, retrieval and gpt4 vision custom function

Primary LanguagePython

LinkedIn Profile Analyzer 🚀💼

Welcome to the LinkedIn Profile Analyzer! This tool not only reads your profile but also visually analyzes your profile picture to provide detailed feedback and actionable suggestions for elevating a LinkedIn profile.

How It Works

Our app leverages the power of OpenAI's GPT-4 and GPT-4 Vision models through the Assistants API, integrating both text and visual analysis capabilities. Here's a quick overview:

  • app.py: The frontend of the application, built with Streamlit. It manages the user interface, facilitating easy interaction and displaying insights directly to you.
  • main.py: Initializes the OpenAI Assistant specifically designed for LinkedIn profile optimization, incorporating retrieval from documents and custom function calling.
  • linkedin_scraper.py: Gathers data from your LinkedIn profile URL, preparing it for comprehensive analysis by the Assistant.

Simple Steps for Users:

  1. Input your LinkedIn profile URL.
  2. The app scrapes your profile data, readying it for in-depth analysis.
  3. Our custom AI Assistant, enhanced with vision capabilities, examines both the content and the profile picture, ensuring a holistic profile evaluation.

Highlights

  • OpenAI Assistants API: Using the Assistants API for textual analysis and recommendations.
  • Custom Function & GPT-4 Vision: A custom function taps into GPT-4 Vision for an insightful critique of profile pictures, evaluating professional presentation and suitability.
  • Retrieval and Function Calling: Seamlessly integrates document retrieval for enriched context and calls custom functions for specialized tasks like image analysis.

Quick Start

  1. Clone and Setup: Clone this repo and install dependencies (pip install -r requirements.txt).
  2. Run the App: Use streamlit run app.py to start the application.
  3. Input and Analyze: Simply enter your OpenAI API key and LinkedIn profile URL in the sidebar, then press "Analyze" to see the magic happen.

Contributing

We welcome contributions! If you have suggestions or improvements, please fork the repo, commit your updates, and submit a pull request.

Tech

  • Streamlit: For creating the web app interface.
  • OpenAI API: Leveraging Assistants API, including GPT-4 for textual analysis and GPT-4 Vision for image analysis.
  • Python: For backend logic and integration with the OpenAI API.

Made with ❤️ and Python.