AI-Powered LinkedIn Job Search & Application Assistant

By voidbydefault

This project is an advanced automation tool designed to streamline the LinkedIn job application process. It utilizes Selenium for browser automation and Google Gemini AI to intelligently parse job descriptions, score relevance, and generate context-aware answers to application questions.

Features

Major Update: Bot is now fully GUI based. No more worrying about managing text-based configurations.

• AI-Driven Application: Uses Google Gemini (Flash model) to evaluate job descriptions against your resume and generate custom answers for specific employer questions.

• Resume Parsing: Automatically extracts professional details from PDF resumes to build a "Source of Truth" user profile.

• Intelligent Filtering: Scores jobs based on compatibility (0-100) and skips low-relevance positions.

• Anti-Detection Mechanisms: Implements undetected-chromedriver, random delays, and human-like interactions to minimize detection risk.

• Analytics Dashboard: Built-in Streamlit dashboard to visualize application success rates, status breakdowns, and geographic data.

• Ban-Safe Mode: Enforces daily application limits to protect account integrity.

Architecture

Prerequisites

• Python: 3.13+

• Google Chrome: Installed locally.

• Gemini API Key: Obtained from Google AI Studio.

Installation

Demo

Watch demo and how-to setup video on YouTube.

Prefer using PyCharm IDE for simpler setup:

1. Download PyCharm IDE

   Download and install from Jetbrains.

2. Add New Python Interpreter

   
   
   
   

3. Configure Virtual Environment

   Select "Generate New", type Virtualenv, and click OK.

   
   
   
   

4. Install Dependencies

   Double click run.py to open it. When you see the notification bar (as shown below), click on sync.

   
   
   
   

Running the Bot

1. Initialize the Script

   Execute the run.py script.

   
   
   
   

2. Complete Configuration

   Once the Google Chrome window appears, use the GUI to complete the setup.

   
   
   
   

Project Structure

├── requirements.txt         # Must-Be-Installed dependencies
├── run.py                   # Entry point: Initializes configuration & launcher
├── app/                     # Main Application Logic
│   ├── ai_handler.py        # AI Core: Gemini integration, Resume parsing
│   ├── config_ui.py         # Configuration Server (Flask)
│   ├── dashboard.py         # Main Analytics Dashboard
│   ├── scout_dashboard.py   # Scout Mode Dashboard
│   ├── defaults.py          # Default AI Rules and Seeds
│   ├── bot/                 # Browser Automation Logic
│   ├── static/              # CSS & Images for UI
│   └── templates/           # HTML Templates for UI
├── config/                  # Configuration Files
│   ├── config.yaml          # Job search parameters
│   ├── secrets.yaml         # Credentials (NEVER SHARE)
│   └── gemini_config.yaml   # AI settings (NEVER SHARE)
├── work/                    # Runtime Data (Logs, Databases, Cache, Resumes)
└── docs/                    # Documentation Assets

Disclaimer

This code comes with no warranties at all, don't blame me if your account is restricted or banned. Keep bot's use fair and reasonable. Don't abuse LinkedIn as a platform or this bot as a tool.

Credits and history

AI-version Linkedin EasyApply AI:

Complete revamp, additional of GUI, and modularization of logics, codebase, enhanced human-like behavior and implementation of GenAI by voidbydefault

Non-AI version EasyAplyBot:

• Grandfather, original developer: Nathan Duma
• Foster dads, maintenance and improvements after Nathan left: Micheal Dingess and voidbydefault

Star History