🔥 Hot Repo: 36K Stars in 15 Days — Dev Turned His Job Hunt Into an Open-Source Weapon

Santiago Fernández de Valderrama spent months as a job seeker fighting AI hiring filters. His answer: build a Claude Code-powered multi-agent system that evaluated 740+ offers and landed him a Head of Applied AI role. Now it has 36,270 GitHub stars and 1,300 Discord members in just two weeks.

By OMC Editorial on 2026-04-19

When Santiago Fernández de Valderrama started his job search, he faced the same AI-filtered gauntlet as every other developer: ATS resume scanners, automated rejection emails, and hiring pipelines designed to process candidates at scale. His response was unusually systematic. He built his own AI system to fight back — and then open-sourced it. career-opshttps://github.com/santifer/career-ops landed on GitHub on April 4, 2026. As of April 19, it has 36,270 stars and 7,318 forks — growth of roughly 2,400 stars per day — making it one of the fastest-growing Claude Code projects since the framework launched. What It Actually Does career-ops is a multi-agent job search pipeline built on Claude Code. It uses Playwright to scrape job portals, Claude to reason against the developer's CV not keyword-match, and a Go-built terminal dashboard to track the application pipeline. The system is local-first: all data stays on the user's machine, sent only to whichever AI provider they configure. The core evaluation loop classifies jobs into six archetypes LLMOps, Agentic, PM, SA, FDE, Transformation, then runs a 6-block analysis: role summary, CV match, level strategy, compensation research, personalization notes, and interview prep using the STAR+Reflection framework. Each job receives an A–F grade weighted across 10 dimensions. The system explicitly discourages applying to anything below 4.0/5 — it is positioned as a filter, not a spray-and-pray tool. Beyond evaluation, it generates ATS-optimized PDFs — keyword-injected CVs styled with Space Grotesk and DM Sans — and accumulates an "Interview Story Bank" of STAR+Reflection answers that improve with each evaluated role. Numbers From The Creator's Own Job Search The README cites specific results from the creator's own use of the system: 740+ job listings evaluated, 100+ personalized CVs generated, 1 role landed — Head of Applied AI. A written case study details the full funnel: 631 evaluations ran before landing the role, w