Auto Apply Project Overview

The "Auto Apply" project automates the entire job application process, from gathering job details to submitting the application.

Core Purpose

To automatically apply for jobs by:

1. Scraping job post information.
2. Generating a custom, optimized resume.
3. Filling out and submitting application forms.

How It Works (Workflow)

The process involves five main steps:

1. Information Extraction: Scrapes the job post URL to extract job details and application form fields.
2. Resume Generation: Creates a customized resume PDF in LaTeX format, tailored for the specific job.
3. Response Generation: Generates answers for the application form, potentially asking clarifying questions.
4. Response Refinement: Evaluates and refines the generated responses based on feedback.
5. Application Submission: Fills out the application form and submits it.

Key Technologies

• LaTeX for Resumes: Utilized for precise formatting and declarative structure, aiding in AI-friendly resume generation.
• Pandoc API: Converts generated LaTeX files into PDF resumes.
• Puppeteer MCP + API: Manages browser automation for scraping and form submission, ensuring separation of concerns and isolated browser environments.

Inputs

• Resume.md (Markdown resume)
• Resume.tex (LaTeX resume)
• Job post url (Link to the job posting)

Output

• Application submitted

To Do List

• Resume generation feedback, critique and gut check from "hiring manager" LLM.
• Robust form parsing and correct fill up.
• LLM response evaluations.

Deploy

See infrastructure deployment guide:

infra/DEPLOYMENT.md