💼 Resume to LinkedIn Formatter (Auto-Generated Profile Builder)

This project automatically extracts sections from an uploaded resume, reformats them into a LinkedIn-style profile using an LLM, and displays the result in a user-friendly Streamlit dashboard.

🧠 Tech Stack

• Frontend: Streamlit
• Backend: AWS Lambda functions (2)
• Database: Amazon RDS (MySQL)
• LLM: Together AI Inference API (mistralai/Mistral-7B-Instruct-v0.1)
• Storage: AWS S3 (for raw uploads, optional)
• Trigger Mechanism: Lambda → Lambda chain via insert event

---

🚀 Features

• Upload a resume (PDF/Text)
• Extract sections like Experience, Education, Projects, etc.
• Format each section for a professional LinkedIn profile
• View the auto-generated LinkedIn sections in Streamlit
• Database-backed pipeline with automatic reformatting

---

1. ✅ IAM Role for Lambda

Make sure your Lambda function has an IAM role with the following permissions:

• AmazonTextractFullAccess
• AmazonS3ReadOnlyAccess
• AWSLambdaVPCAccessExecutionRole (if accessing RDS in a VPC)

MySQL

• Create the following tables: 'resume_sections' and 'linkedin_profile_sections'

CREATE TABLE resume_sections ( id INT AUTO_INCREMENT PRIMARY KEY, section VARCHAR(255), content TEXT, created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP );

CREATE TABLE linkedin_profile_sections ( id INT AUTO_INCREMENT PRIMARY KEY, section VARCHAR(100), content TEXT, created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP );

Install Dependencies

• pip install -r requirements.txt

Setting Up the .env File

• Replace all the placeholder texts for each key in the .env file with the corresponding information

How to Run The Project

• python3 -m venv venv
• source venv/bin/activate
• cd resume-to-linkedin
• streamlit run app.py --server.port 8501 --server.address 0.0.0.0