hiring-agent

Why we included this project

Hiring teams facing a flood of entry-level applications will get real mileage out of this one. It turns resume PDFs into structured JSON, pulls in GitHub profile and repository signals, and returns an explainable score with per-category evidence, bonus points, and deductions. You can run it entirely locally through Ollama or point it at a hosted Gemini backend, and the CLI plus Jinja prompt templates are there to read and tweak. What stands out is the honesty around the whole thing: the scoring rubric is public, the default cutoff is deliberately low so only the very bottom of the distribution gets filtered, and the authors say plainly that this ranks resumes to help humans decide what to read first, not to replace the decision. There's also useful coverage of score variance and the risk of prompt injection through hidden PDF text, which makes it a solid starting point if you're building your own explainable evaluation pipeline.