AI Job Search FAQ

AI Job Search is an AI-powered job application framework built on Claude Code. It automates your entire job search process by evaluating job postings, tailoring your CV, writing cover letters, and preparing you for interviews. The system runs locally on your machine, giving you full control over your data while leveraging AI to create highly personalized applications.

The framework uses a structured workflow with three core commands:

• /setup - Creates your detailed profile from documents, CV imports, or interviews
• /scrape - Searches job portals and presents matches with fit ratings
• /apply - Evaluates fit, drafts tailored documents, reviews with a second agent, and verifies PDF output

The system includes advanced features like PDF verification loops, ATS compatibility checks, and drafter-reviewer workflows to ensure high-quality output.

Before starting, you'll need:

• Claude Code (CLI) - The core AI assistant
• Python 3.10+ - Runtime environment
Bun - For job search CLI tools
• LaTeX distribution with lualatex and xelatex (TeX Live, MacTeX, TinyTeX, or MiKTeX)
• Optional: pdftotext from poppler for ATS compatibility checks

Unlike generic AI tools, AI Job Search includes:

• PDF Verification Loop - Compiles and visually insures every PDF until layout is perfect
• ATS Verification - Checks documents the way Applicant Tracking Systems see them
• Smart CV Cutting - Relevance-weighted trimming when CVs exceed 2 pages
• Drafter-Reviewer - Separate AI agent reviews drafts and provides honest feedback
• Security First - Agentic defenses against untrusted job postings

Three main customization options:

• Custom Templates - Use /add-template to register your own CV or cover letter templates (LaTeX, Typst, etc.)
• Job Portals - Use /add-portal to generate search skills for local job boards
• Profile Updates - Edit your profile files directly or use /setup --section search to update job search preferences

Absolutely! While the shipped tools focus on Danish job portals, the framework is designed for global use:

• LinkedIn Search - Works in any country with LinkedIn public job listings
• Freehire Search - Multi-market tech job aggregator
• Custom Portals - Use /add-portal to create skills for your local job boards

The core evaluation and document generation workflows are completely language- and country-agnostic.

The salary tool works with any data you provide:

• Import union statistics, Glassdoor exports, or personal research
• Convert Excel data to JSON format using provided tools
• Use during application process for salary negotiation preparation
• Optional feature - skipped if no salary data is provided

No, this is an independent open-source project. While it uses Claude Code as its AI foundation, it is not affiliated with, endorsed by, sponsored by, or maintained by Anthropic. The project has no cryptocurrency, token, or paid sponsorship programs - any claims to the contrary are unauthorized and should be treated as scams.

Use these practices to stay current:

• Update to tagged releases rather than raw master
• Use python3 tools/check_upstream_updates.py to preview changes before merging
• Run python3 tools/upstream_triage.py to prioritize which commits to review
• Follow the detailed update workflow in SETUP.md section 8

Contributions are welcome! First read CONTRIBUTING.md which explains:

• What types of contributions get merged
• What belongs in forks vs. upstream
• Quality standards for code and documentation
• Testing requirements
• Community guidelines

Common contributions include new portal skills, template improvements, and bug fixes.