How the Resume Scorer Grades a Resume (Heuristics, Not an LLM or ATS)
The Resume Scorer is a rule-based analyzer that runs entirely in the browser with no AI, LLM, server or network call, and it is not a real applicant-tracking system. It accepts either pasted resume text or an imported resume-builder JSON file (normalized and rendered to plain text via the shared resume engine), then splits the text into achievement bullets and scores seven weighted writing-quality dimensions drawn from recruiter and university career-office guidance. Impact & metrics measures the share of bullets containing a measurable result (a number, %, $ or quantity word). Action verbs measures the share of bullets that open with a strong verb from a curated list (Led, Built, Increased, Reduced). Clichés & weak words counts overused or passive phrases (“responsible for”, “duties included”, “team player”, “results-driven”). Key sections keyword-detects contact details, summary, experience, education and skills. Conciseness checks the average bullet length against ~15–25 words. Length checks the total word count against the ~400–800-word (1–2 page) band. The final check flags first-person pronouns. Each dimension is marked Pass, Warn or Fail with a reason and a fix, and the weighted points are summed to a transparent score out of 100.
- Rule-based heuristics built on the shared resumeEngine helpers — no AI, LLM, server or ATS
- Seven dimensions: Impact & metrics, Action verbs, Clichés/weak words, Sections, Conciseness, Length, First person
- Scores pasted resume text OR an imported resume-builder JSON file
- Each dimension marked Pass / Warn / Fail with a reason and a concrete fix
- Weighted checklist tally produces a transparent score out of 100
