ai-job-search: an AI-powered job application framework that automates tailoring, PDF layout verification, and ATS-compatibility checks for resumes
ai-job-search: an AI-powered job application framework that automates tailoring, PDF layout verification, and ATS-compatibility checks for resumes
What it solves
This project provides an automated framework for managing the end-to-end job application process. It eliminates the manual effort of searching for jobs, tailoring resumes and cover letters for every single application, and preparing for interviews, while ensuring that the final documents are professionally formatted and ATS-compatible.
How it works
Built on Claude Code (CLI), the system uses a structured workflow to turn the AI into a career assistant. It begins with a /setup phase where it builds a comprehensive candidate profile from existing documents or interviews. Users can then /scrape job portals (using built-in CLI tools for specific markets or custom ones created via /add-portal) and /rank them based on fit.
The /apply command executes a "drafter-reviewer" pipeline: one agent drafts a tailored CV and cover letter in LaTeX, and a second agent critiques the draft based on company research. The system then compiles the LaTeX into PDFs and visually inspects the layout to fix orphans or page overflows. Finally, it uses pdftotext to verify that the PDF's text layer is readable by Applicant Tracking Systems (ATS) before presenting the final files.
Who it’s for
Job seekers who want to automate the tedious parts of applying for roles, particularly those comfortable with CLI tools and LaTeX for high-quality document production.
Highlights
- PDF Verification Loop: Automatically iterates on LaTeX code until the rendered PDF meets strict page limits and layout standards.
- ATS-Check: Extracts the PDF text layer to ensure contact details and keywords are correctly parsed by automated systems.
- Relevance-Weighted Cutting: When a CV is too long, it removes the least relevant experience based on the job posting rather than just cutting the oldest entries.
- Drafter-Reviewer Architecture: Uses two separate AI agents to ensure drafts are critiqued and researched before finalization.
- Interview Prep: Generates stage-specific prep packs and mock interviews based on the specific CV and cover letter submitted for that role.
Sources
- undefinedMadsLorentzen/ai-job-search