LinkedIn Post Automation

Overview

LinkedIn Post Automation is an application designed to automate the process of creating, and posting content on LinkedIn. Leveraging AI-powered tools for content generation, this project enables users to optimize their LinkedIn posts for maximum engagement and professionalism.

Architecture

Features

1. AI-Powered Content Generation: Automatically generate professional LinkedIn posts tailored to your input.
2. Automated Posting: Schedule and post content directly to LinkedIn for multiple days.

Tech Stack

Frontend

• Framework: NextJs
• Styling: Tailwind CSS

Backend

• Framework: FastAPI, Redis, Dramatiq
• Languages: Python

Project Flow

The LinkedIn Post Automation system automates the entire process of content creation, and posting to LinkedIn. Here’s how it works:

1. User Input and Content Generation

• Frontend: The user provides a LinkedIn post topic (text-based prompt) on the NextJs interface.
• System Response:
  • The frontend sends a request to the backend to generate the LinkedIn post content and custom image.
  • The backend calls the AI model APIs (GROQ, Gemini) to generate the content based on the input topic.
  • Similarly, the backend generates a custom image using AI models.
  • The frontend receives the generated content (text) and image (as PNG).

2. Posting to LinkedIn

• Backend Process:
  • The backend uses the LinkedIn API to upload the image and post the content.
  • The process includes two main steps:
    • Image Upload:
      • The backend uploads the image to LinkedIn using the POST /assets endpoint of the LinkedIn API. This creates a new image asset on LinkedIn.
      • The API responds with an asset ID (a unique identifier for the image).
    • Post Creation:
      • The backend then uses the LinkedIn API's "Share on LinkedIn" endpoint to create a post. The request includes the post content and the image asset ID.
      • The backend successfully creates the post and returns a confirmation message with the LinkedIn post details (URL, post ID).

• System Response: The frontend shows a success notification, confirming that the post has been successfully uploaded to LinkedIn.

Summary of Flow:

1. Frontend: User inputs a post topic description → Sends request to backend.
2. Backend:
   • Generates post content (AI-powered).
3. Post Creation: Backend creates a LinkedIn post using the content asset ID → Post is uploaded to LinkedIn.

Setup Guide

Prerequisites

• Node.js (>= 16.x)
• Python (>= 3.8)
• Redis Insights

Frontend

1. Navigate to the client folder:

   cd client

2. Install dependencies:

   npm install

3. Start the development server:

   npm run dev

Backend

1. Create a virtual environment and activate it:

   uv venv .venv
   .venv/Scripts/activate

2. Install dependencies:

   uv pip install -r requirements.txt

3. Create a .env file and add the following keys:

    ACCESS_TOKEN=<access_token>
    EMAIL=<email_id>
    GROQ_API_KEY=<groq_key>
    HUGGINGFACE_API_KEY=<huggingface_key>
    PASSWORD=<password>
    PERSON_URN_KEY=<urn_key>
    TAVILY_KEY=<tavily_key>
    REDIS_URL=<redis_url>

4. Build the docker-compose:

   docker-compose build

5. Run the docker-compose file:

   docker-compose up

Contribution

Feel free to contribute by:

• Reporting issues
• Suggesting features
• Submitting pull requests

License

This project is licensed under the MIT License.