cd /news/ai-agents/claude-code-job-search-69-applicatio… · home topics ai-agents article
[ARTICLE · art-116579] src=dev.to ↗ pub= topic=ai-agents verified=true sentiment=↑ positive

Claude Code Job Search: 69 Applications, 20 Interviews, 1 Offer

Mads Lorentzen, a laid-off geophysicist, used Claude Code to build an open-source agentic job-search framework that automated his entire job hunt, resulting in 69 applications, 20 interviews, and a job offer as an AI engineer. The project, ai-job-search, has gained over 29,000 GitHub stars and includes features like dual-agent resume review, ATS-verified LaTeX documents, and local-first privacy. Lorentzen also credits AI coding tools for his markdown editor ColaMD 2.0.0, which has surpassed 1,000 stars and 1,500 installs.

read2 min views3 publishedAug 31, 2026

A laid-off geophysicist turned his job hunt into an agentic workflow and open-sourced it. The project, ai-job-search, has blown past 29,000 stars on GitHub — and it's a blueprint for how AI turns a messy, subjective process into an executable pipeline.

From 69 applications to an offer (source: @aigc1024). Mads Lorentzen, a geophysicist who was laid off, used Claude Code to build a job-search framework that automated his entire funnel: 69 applications → 20 first-round interviews → a job offer, and he landed an AI engineering role. The full funnel data is shared publicly on LinkedIn.

The whole job-search supply chain, automated (source: @aigc1024). Where most people use ChatGPT as a copy-paste resume polisher, ai-job-search rebuilds the pipeline end to end: job discovery, match scoring (skills, experience, culture, location, career growth), tailored resume and cover letter drafting, interview prep, and application tracking.

Dual-agent review is the core design (source: @aigc1024). A Drafter agent generates a tailored resume and cover letter from your profile; a Reviewer agent — a second Claude agent started with fresh context — independently researches the target company and critiques the draft from a third-party angle. That directly attacks the quality instability of single-pass generation.

Four commands cover the full loop (source: @aigc1024). /setup

builds your career profile (from a documents folder, a single resume, or a guided interview); /scrape

searches and ranks jobs (built-in Danish portals, plus /add-portal

to extend to any region); /apply generates materials — LaTeX-compiled to an exactly-2-page resume and 1-page cover letter, with ATS verification on the PDF text layer; /interview

produces staged interview prep with company research, interviewer background, STAR case mapping, and mock interviews.

Local-first, privacy-safe, fork-and-own (source: @aigc1024). All data stays on your machine — no SaaS, no privacy leaks. Fork it and change anything.

Bonus: ColaMD 2.0.0, a markdown editor shipped by a self-taught non-programmer (source: @aigc1024). The author credits AI coding tools — specifically the ox and terra models — for pushing each version forward, including a notoriously harder iOS build. Now at 1,000+ stars and 1,500+ installs, with Mermaid support, multi-window, custom fonts, and auto-save.

The pattern is the point: complex, subjective workflows become repeatable agent pipelines — and the people shipping those pipelines are the ones getting hired.

── more in #ai-agents 4 stories · sorted by recency
── more on @mads lorentzen 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/claude-code-job-sear…] indexed:0 read:2min 2026-08-31 ·