Show HN: Shobr: Job seach CLI via browser automation, event-sourcing, LLMs (github.com)

SHOBR, a new job search automation tool, has been launched, offering a unique approach that combines browser automation, event sourcing, and minimal use of large language models (LLMs). Unlike other tools that utilize headless browsers or require users to share their credentials, SHOBR operates through the user's existing authenticated browser using the beachpatrol framework, providing a stealthy scraping technique that mimics regular user activity. The system employs a human-in-the-loop model, where SHOBR drafts applications and generates tailored resumes in Markdown format—a method that ensures final submissions are always manually verified by the user. This tool is significant for the AI/ML community as it presents a more transparent approach to job automation, requiring minimal LLM interaction while remaining agnostic to various LLM providers. SHOBR leverages an event-sourced data structure, allowing users to track and manage job applications efficiently without data loss. The use of a Markdown-based CV toolchain integrated with Groff and Pandoc ensures that resumes are ATS-ready while maintaining a straightforward and customizable setup through a well-structured command line interface. Overall, SHOBR aims to enhance the job search process, making it less of a "spray and pray" approach and more systematic and tailored to individual user needs.