GitHub radar
AI Job Search: Claude Code Framework for Job Hunting
A workflow built on top of Claude Code that helps you discover jobs, calculate fit scores, generate personalized resumes and cover letters, and interview prep. (42,000 GitHub Stars)
This is a free and open-source job application framework built on top of Claude Code. You only need to enter your profile information once and then you can use some commands to help you apply for a job. Use /scrape to search online for job listings and see recommendations for jobs you will be a good fit for along with a fit score. Use /apply to assess a job, write a personalized resume and cover letter with a recommendation from a reviewer agent and then send the final product. Use /interview to assist you in interview prep. This tool was developed by geophysicist Mads Lorentzen who lost his job late 2025. He used this framework to apply for new jobs and found a job as an AI engineer in June 2026 after 69 personalized applications, 20 first round interviews, and 1 signed contract. This framework is currently language agnostic and country agnostic, you can also integrate this framework into other agents like Codex or Gemini CLI using AGENTS.md. This is approximately 42,000 GitHub stars, and is under a MIT license. This is an independent project and is not associated with Anthropic, the creators of Claude Code.
Why a vibe-coder should care
Writing a personalized application for each job is very time consuming. With this agent, you can get a first draft of the resume and cover letter for each job with another agent reviewing it, saving you time to select jobs to apply to and interview prep.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Set up AI Job Search from https://github.com/MadsLorentzen/ai-job-search — make a private copy as the README recommends, install the prerequisites, run /setup and help me fill in my profile. Ask me if you need any access.
Runs on a regular laptop — you'll need Claude Code, Python, Bun, and LaTeX to build the résumé as a PDF.
Open on GitHub▌ More finds