🔍 Job Search Automation

Automated job scraper that runs every weekday and updates jobs/jobs.json with fresh remote, Berlin-based, React/Next.js roles — filtered for English-only listings and visa sponsorship where available.

Sources

Platform	Method	Visa Filter	Salary
Arbeitnow	Public API ✅	✅ Yes	❌
Wellfound	HTML scrape	Partial	✅ Yes
EnglishJobs.de	HTML scrape	❌	❌
LinkedIn	RSS feed	❌	❌

Setup

1. Fork / clone this repo to your GitHub account

2. Enable GitHub Actions

Go to your repo → Actions tab → Enable workflows.

3. No secrets needed!

All sources are scraped publicly. No API keys required.

4. Run manually anytime

Go to Actions → 🔍 Daily Job Search → Run workflow.

Schedule

Runs Monday–Friday at 9:00 AM Berlin time (8:00 UTC).

Output

All jobs are saved to jobs/jobs.json with this structure:

[
  {
    "id": "arbeitnow-senior-react-developer-berlin",
    "source": "Arbeitnow",
    "title": "Senior React Developer",
    "company": "Acme GmbH",
    "location": "Berlin (Remote)",
    "remote": true,
    "visa_sponsorship": true,
    "salary": null,
    "url": "https://...",
    "tags": ["react", "typescript", "remote"],
    "posted_at": "2025-03-01",
    "fetched_at": "2025-03-02"
  }
]

Run locally

npm run fetch

Filters applied

• Keywords: react, next.js, nextjs, frontend, front-end
• Location: Berlin / Remote
• Deduplication: Jobs are never added twice (tracked by id)
• History: Keeps up to 500 most recent listings

Updating resume data

When you update the resume PDF at public/Lucas-Johnson-Frontend-Engineer-Resume.pdf, run:

pnpm update-resume

This invokes Claude Code to read the PDF and update:

• app/lib/resumeData.ts — skills, experience, education, and summary
• app/lib/docgen/coverLetter.ts — fit points and opening paragraph

You can also pass a custom PDF path:

pnpm update-resume -- path/to/resume.pdf