ResumeParser.pro

Score Any Resume Against Any Job — Across 20 Dimensions

One number can rank candidates. Twenty numbers with written reasons can defend the ranking to a recruiter, a candidate, or an auditor. That difference is the product.

Start scoring free Endpoint docs

14-day trial · 100,000 words included · No credit card

The SharpAPI Resume Job Match Score API takes a resume file and a job description and returns a 0–100 score across 20 dimensions (skills, experience, education, technical stack, seniority, stability and more), each with a plain-language explanation of the evidence.

A real response, dimension by dimension

This is the sample payload for a senior PHP developer scored against a Laravel role. Note the zeros: the resume has no education section, so the scorer says so instead of guessing.

overall_match 72 Weighted aggregate of all 20 dimensions
skills_match 80 Strong PHP/MySQL; Laravel not mentioned explicitly
experience_match 90 22 years of programming — exceeds requirements
education_match 0 No education section found in the resume
technical_stack_match 80 Backend stack aligns with the posting
job_title_relevance 70 Titles adjacent to the advertised role
industry_experience_match 85 Long record in software product companies
language_proficiency_match 100 Fluent in the posting’s required language
management_experience_match 100 Led teams at the required level
remote_work_flexibility 90 Documented remote work history
stability_score 85 Healthy tenure lengths, no job-hopping pattern
location_preference_match 20 Candidate is in a different region than the role

Sample response · 12 of 20 dimensions shown · also scored: certifications, project experience, methodologies, soft skills, location preference, training relevance, years-of-experience weighting, recency of relevant roles

How the weighting works

Not all dimensions matter equally, and the rubric is public rather than secret:

Requirements the job description marks as must-have weigh three times more than nice-to-haves. And when your pipeline knows better, say so: the context parameter takes up to 5,000 characters of plain-text instructions — emphasize a certification, de-emphasize location, credit adjacent experience — and the weighting adjusts.

The request

POST /api/v1/hr/resume_job_match_score
curl -X POST 'https://sharpapi.com/api/v1/hr/resume_job_match_score' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -F 'file=@candidate.pdf' \
  -F 'content=We are hiring a Senior Backend Developer (PHP/Laravel)…' \
  -F 'language=English' \
  -F 'context=EMPHASIZE: AWS certification. DEEMPHASIZE: on-site availability.'

Same async pattern as every SharpAPI endpoint: 202 Accepted, a status URL, poll or webhook. Score two candidates or two thousand — the calls run in parallel.

Scoring that survives an audit

Before scoring, the pipeline strips personally identifying information from the parsed resume: name, email, phone, address, date of birth, nationality, photo and profile links. The model evaluates professional qualifications and nothing else.

Combined with per-dimension explanations, that design maps directly onto what regulators now ask of hiring AI — bias audits under NYC Local Law 144 and transparency duties under the EU AI Act. A score you can explain is a score you can keep using.

Where teams point it

ATS shortlisting — recruiters open a ranked list, not a pile of PDFs. 200 applicants → sorted by overall_match

Job boards — show applicants a fit score before they apply; fewer spray-and-pray applications, better matches. candidate ↔ posting → fit badge

Staffing agencies — match the whole candidate bench against a new client brief in one batch run. bench × brief → ranked longlist

Career tools — show job seekers which requirements they miss and what to fix. gap report → coaching plan

Questions, answered

What does the Resume Job Match Score API return?

A 0-100 score for each of 20 dimensions — skills, experience, education, certifications, technical stack, job-title relevance, industry experience, soft skills, language proficiency, management experience, stability and more — plus an overall_match weighted aggregate and plain-language explanations that name the evidence behind the key scores.

What inputs does the API need?

Two things: the resume file (PDF, DOCX, RTF, TXT or image) and the job description as plain text. Optionally add a language for the explanations and a context string of up to 5,000 characters that tells the scorer what to emphasize, de-emphasize or credit.

How is the match score calculated?

Dimensions carry weights: skills, experience and technical stack count triple; job-title relevance, industry experience, recency, language and years of experience count double; the rest count once. Requirements marked as must-have in the job description weigh three times more than nice-to-haves. Every headline score ships with a written reason, so it is never a black box.

Can it score a resume in one language against a job ad in another?

Yes. A German CV can be scored against an English job description, and the explanations come back in whichever of 80+ languages you request.

Does the score discriminate by name, age or nationality?

Personally identifying information — name, email, photo links, age, nationality, social profiles — is stripped from the parsed resume before scoring. The model evaluates professional qualifications only, which supports bias-audit requirements like NYC Local Law 144.

What does it cost to score candidates?

Per-word metered pricing from $50/month, same as all SharpAPI endpoints. The 14-day trial includes 100,000 words — enough to score a real applicant pool — with no credit card required.

Further reading