AI Resume Judge

Problem

Resume screening often relies on gut feel. A resume can look acceptable while weak in evidence: vague responsibilities, unquantified impact, flat skill lists, and generic self-assessment make it hard to place a candidate against a target role. This skill turns resume review into a checkable rubric, with a stronger focus on AI and large-model roles and an optional cross-industry reference score.

How it works

It accepts resume text in Markdown or a file path, parses the content, and detects whether the profile appears AI-related. Scoring uses a 100-point breakdown: basic information completeness 15, quantified impact 25, technical depth and breadth 20, work experience quality 20, education 8, personal brand or influence 7, and communication clarity 5. It applies deduction rules for vague verbs such as assisted or participated, missing numbers, unlayered skills, and boilerplate self-introductions. Non-AI resumes receive a raw capability score first, then an equivalent reference score using industry coefficients, such as 0.72 for DevOps.

Boundaries

This is better suited to early screening, feedback, and revision suggestions than to hiring decisions. It can only judge what is written in the resume; it cannot verify project ownership, code quality, collaboration, or role fit. Cross-industry coefficients are heuristics, not proof that a candidate is interchangeable across domains.