Dev codes AI job matcher with nuxt 4 after 200 auto-applies flop: because scraping naukri and claude haiku beats begging recruiters for mercy

A developer, frustrated with the ineffectiveness of auto-applied job applications, has created an AI-powered job matching system that prioritizes human judgment. After sending over 200 auto-applied job applications with no responses, the developer built a job portal using Nuxt 4, Vue 3, and TypeScript, which utilizes a weighted scoring system to match job seekers with relevant opportunities. The system, powered by Claude Haiku AI, evaluates each job based on factors such as tech stack, role alignment, and experience level, with a score of 50 or higher indicating a strong match. The developer's approach has yielded positive results, with 4 interviews from 15 manual applications, compared to no interviews from over 200 auto-applied applications. The open-source project is available on GitHub, with the developer seeking contributors to enhance the platform with additional job sources and features, such as resume tailoring suggestions and interview prep notes.