LinkedIn LLM Post Generator

🚀 An AI-powered LinkedIn Post Caption Generator that leverages Large Language Models (LLMs) to create professional, creative, and engaging post captions. This project combines LinkedIn post data scraping with LLM-based text generation.

---

📌 Problem Statement

Creating impactful LinkedIn posts consistently is time-consuming and requires creativity. Many professionals struggle to write captions that are engaging, audience-focused, and tailored to LinkedIn’s professional environment.

This project solves that problem by scraping real LinkedIn post data and using LLM models (Gemini API & GPT-2) to automatically generate effective LinkedIn post captions.

---

🛠️ Project Workflow

1. Scraping Data

   • Linkedin-post-Scrapper.py scrapes LinkedIn posts and stores them in a structured dataset (linkedin_data.csv).

2. Dataset

   • linkedin_data.csv contains scraped LinkedIn post text used for training and fine-tuning.

3. LLM Models for Caption Generation

   • llm_gemini_api.py → Uses Google Gemini API to generate captions.
   • llm_gpt2.py → Uses Hugging Face’s GPT-2 model to generate captions.

4. Results & Example

   • Sample-input-output.png shows sample input prompts and generated LinkedIn captions.

---

📂 Project Files

• Linkedin-post-Scrapper.py → Scrapes LinkedIn posts into a CSV dataset.
• linkedin_data.csv → Scraped LinkedIn data used as training/analysis input.
• llm_gemini_api.py → Caption generation using Google Gemini API.
• llm_gpt2.py → Caption generation using GPT-2 model.
• Sample-input-output.png → Example of generated LinkedIn captions.
• README.md → Project documentation.

---

⚙️ How to Run

1. Clone this repository

   git clone https://github.com/Chandrashekar0123/linkedin-llm-postgen.git
   cd linkedin-llm-postgen
   

🚀 Future Improvements

Fine-tune LLM models on larger LinkedIn datasets.

Add sentiment/emotion-based caption generation.

Deploy via Streamlit/Flask for interactive UI.