Eight weighted dimensions calibrated to your career stage, every deduction named, and the arithmetic shown in front of you. The rubric is published and versioned, so the number means the same thing next month as it does today.
At a glance
Each report costs 2 credits. A new verified account gets 20, plus one free deterministic run every day, forever - and re-running an unchanged resume never costs anything. No subscription and no card.
The run needs an account because it spends credits, and credits are what pay for the AI call. The free ATS checker beside it needs no account at all and never will - it makes no external call, so it costs us nothing to run.
The AI in this pipeline is handed findings our code has already computed and asked to explain them. It is never asked to score anything, and it structurally cannot - the schema its reply is validated against contains no numeric field at all.
The number
8 dimensions are each scored 0–100 in our own code, multiplied by a published weight, and added up. The report shows every sub-score, every weight and every contribution - and they sum to the total in front of you. You can check it with a calculator, and people do.
Reproducible
No model randomness and no dependence on the date or the order findings arrived in. Run the identical resume twice and the number does not move - which is not true of a scorer that asks a language model for it.
No invention
Where a bullet needs a figure your resume does not contain, the rewrite inserts a marked placeholder and lists it separately. A fabricated number you paste into a real resume is an interview you fail.
Redacted first
Name, email, phone, address and date of birth are replaced with tokens before anything leaves our servers. The report you read is written about [CANDIDATE] - the sample below is not censored, that is genuinely what the pipeline sees.
Free to try
Ten complete reports before you pay anything, plus one free deterministic run every day forever, and re-running an unchanged resume never costs a credit. No subscription, no card at signup.
No brand names in this table on purpose. Prices and feature lists in this category move every few months, and a stale figure in somebody else’s column would undermine the one thing this page is actually selling - that our numbers are checkable. So the columns are approaches, and every claim in ours is verifiable in the sample below.
The third column is not a consolation prize. A tool can tell you a bullet has no number in it. It cannot decide which of your projects is worth two lines and which is worth six. If that is the question you actually have, the review is the honest answer.
Resume Review from ₹299This is a sample report on invented data, shown so you can see exactly what you get before spending a credit. No real resume, and no real person. Everything inside the window below is the real component the dashboard renders - not a screenshot and not a trimmed version of it.
Revquix Readiness Score · rubric r1
up 11 points from your last run5 days ago
This is a good general resume. What is left is refinement, and the highest-value version of that is tailoring it to a specific role rather than polishing it further.
In one line: Nine of your fourteen bullets describe what you built without saying what changed because of it - fixing that one pattern is worth more than every other finding on this report combined.
Your fixes, in the order to make them
Ranked by what each fix is worth against how long it takes, so the quick wins come before the ones that need you to go and find a number. Open a fix for why it matters and exactly what to change.
5 fixes worth +25.1 points. Tick each one off as you make it.
No Summary, Profile or Objective heading was found. The document goes straight into its sections without telling a reader what you do.
Why it matters: The summary is the only part of a resume a human reliably reads in full, and it is what tells them which of the roles below to care about. Without it a reviewer has to infer your positioning from your most recent job title, which is frequently the wrong answer for anybody changing direction.
How to fix it: Three or four lines at the top: what you do, how long you have done it, and the kind of role you are looking for. Write it last, after the rest of the resume is final.
Minutes. A mechanical edit - you can do this before you close this tab.
2 section headings did not match a label parsers look for, alongside 4 that did.
2 section headings did not match a label parsers look for, alongside 4 that did.
What I Bring · Beyond Work
Why it matters: Parsers segment a resume by matching headings against a known list, so a creative label is read as body text and folds the whole section into whatever came before it. Both of yours sit above content you presumably want indexed separately.
How to fix it: Rename them to the plain words - "Skills" and "Interests". Keep the creativity for the bullets, where a human will actually read it.
Minutes. A mechanical edit - you can do this before you close this tab.
4 clichés across 3 distinct phrases. Each one is a claim that would be equally true of a candidate you would not hire, which means it is taking space and giving nothing back.
4 clichés across 3 distinct phrases. Each one is a claim that would be equally true of a candidate you would not hire, which means it is taking space and giving nothing back.
results-driven · team player · passionate about
Why it matters: A screener has read each of these several hundred times, so they carry no information - but they do carry a signal, which is that the resume was assembled from a template rather than written. That is a worse impression than saying nothing in the same space.
How to fix it: Delete all four. If the sentence still says something afterwards, it was fine; if it becomes empty, that sentence was never doing any work.
Minutes. A mechanical edit - you can do this before you close this tab.
5 skills appear in this resume without a bullet showing you using them, which reads as familiarity rather than experience - and it is the gap an interviewer opens with.
5 skills appear in this resume without a bullet showing you using them, which reads as familiarity rather than experience - and it is the gap an interviewer opens with.
React · Docker · GraphQL · Terraform · Redis
Why it matters: These five appear in your skills list and nowhere else in the document. To a reader that is a claim with nothing behind it, and to an interviewer it is a list of things to probe - "you mention React, tell me about something you built with it" is the most common opening question there is.
How to fix it: For each of the five, either attach it to a bullet that shows you using it on something real, or remove it. A shorter list you can defend beats a longer one you cannot.
Under an hour. Real editing, but nothing you need to go and find out.
Recruiters scan for numbers, and 79% of your bullets contain none. "Improved performance" is invisible next to "cut p95 latency 40%", and the difference is one figure you already know.
Recruiters scan for numbers, and 79% of your bullets contain none. "Improved performance" is invisible next to "cut p95 latency 40%", and the difference is one figure you already know.
Worked on the payments team and helped improve the system.
Why it matters: A screener comparing forty resumes is looking for a reason to keep reading past the second bullet, and a number is the fastest one there is. Without it your work reads as the same work everybody else in the pile describes - which is not a judgement about the work, only about the sentence.
How to fix it: Take your five strongest bullets and add one figure each to the result half: a percentage, a rupee amount, a latency, a count, a time saved. You do not need precision - "roughly a third" is a number and "significantly" is not.
Needs facts you don't have to hand. You will need a figure, a document or an answer from somebody else before you can write this.
Copy and adapt
Your own bullets, sharpened by the AI from your words alone. It never scores anything and never sees your name, email or phone number. Read each one before you paste it: check the verb still describes what you actually did.
Some rewrites below contain figures we inferred, not figures you gave us. Every one is marked. Replace them with your real numbers before using these - we will never present an invented number as your own, and neither should your resume.
Worked on the payments team and helped improve the system.
Rebuilt the payments retry pipeline, cutting failed transactions [X%] and recovering roughly [₹Y] per month in lost settlements.
Replace before using
The figures marked below were inferred to show the shape of a strong bullet. Replace them with your real numbers before using this - we never present an invented figure as your own.
Responsible for daily standups and sprint planning for the team.
Ran sprint planning and standups for a team of [N], holding [X] consecutive sprints to committed scope.
Replace before using
The figures marked below were inferred to show the shape of a strong bullet. Replace them with your real numbers before using this - we never present an invented figure as your own.
Migrated the reporting service to a new database, reducing query time by 60%.
Migrated the reporting service from MySQL to PostgreSQL, cutting median query time 60% and removing the nightly batch window entirely.
No invented figures. Every number here is from your own bullet - the wording is still yours to check.
4 runs · +14 overall
Oldest run a month ago, newest today.
The arithmetic
This score is arithmetic, not a judgement call by an AI. Every dimension below is measured in our own code, weighted as published, and added up - the contributions sum to the total, so you can check it yourself. The weighting is calibrated to your career stage, and the stage we read is shown beside the score. The AI writes the explanations and the bullet rewrites; it never produces a number and is never asked to re-judge a finding.
Whether an applicant tracking system can read your file at all - layout, fonts, encryption, and whether your contact details are in the text flow.
weight 18%contributes 16.6 to your score1 finding in Parseability, jump to the first
Whether the sections a recruiter and a parser both expect are present, complete and in a conventional order.
weight 12%contributes 8.4 to your score1 finding in Structure & completeness, jump to the first
Whether your bullets state outcomes with numbers, or only describe duties.
weight 21%contributes 11.6 to your score1 finding in Impact & quantification, jump to the first
Whether the skills you claim are backed by a bullet that demonstrates them - matched by meaning rather than by keyword, so a skill can be present in your resume and still count as unevidenced.
weight 17%contributes 11.6 to your score1 finding in Skill coverage, jump to the first
Whether your dates are consistent, in order, and free of unexplained gaps.
weight 8%contributes 7.0 to your score
Spelling, passive voice, cliches, first-person pronouns and bullet length - the things that make a resume read as filler.
weight 7%contributes 6.3 to your score
Whether your resume shows a career going somewhere - progression, tenure that holds, and the size of what you have owned.
weight 10%contributes 6.8 to your score1 finding in Career trajectory & scope, jump to the first
Full marks in 1 dimension
The 8 contributions add up to 75.3, which rounds to your score of 75. Each one is that dimension's sub-score multiplied by its published weight.
Claimed versus demonstrated
Matched by meaning rather than by keyword, so a skill can be found in your resume and still count as unevidenced. Only skills a bullet demonstrates are treated as evidence.
12 evidenced5 mentioned only
Strongly evidenced12
Backed by a bullet that shows you doing something with it.
Marked “via” means no bullet names that skill, but a bullet shows something that cannot be done without it. We count those, because an interviewer would.
Mentioned, not evidenced5
Present in your resume, but no bullet demonstrates it. This is the gap an interviewer opens with.
Expected for "Backend Engineer" and not evidenced here5
Drawn from the skill area your role belongs to, so everything here is the same discipline as the job you named. A study list, not a verdict - nobody has every skill on a job description. If you do have one of these, the problem is that your resume does not say so.
Check any finding yourself
Every finding above was measured against this text, not against your file's visual layout. If something looks wrong in the report, this is where you find out why.
Your email and phone show as tokens because we do not keep them - they parsed correctly. Everything else is verbatim.
[CANDIDATE]
[EMAIL] | [PHONE] | Bengaluru, India | [URL]
EXPERIENCE
Senior Backend Engineer, Payments
Fintech Co · Mar 2023 – Present
• Migrated the reporting service from MySQL to PostgreSQL, reducing query time by 60%
• Worked on the payments team and helped improve the system
• Responsible for daily standups and sprint planning for the team
• Built and shipped the settlement reconciliation job used by finance every morning
Backend Engineer
Product Co · Jul 2021 – Feb 2023
• Developed REST APIs consumed by the Android and iOS clients
• Added integration tests across the ordering service
• Reduced deployment time from 40 minutes to 6 by rewriting the pipeline
WHAT I BRING
Java, Spring Boot, PostgreSQL, Kafka, React, Docker, GraphQL, Terraform, Redis, Git, Linux
EDUCATION
B.E. Computer Science
Engineering College · 2017 – 2021
BEYOND WORK
Open-source contributor, competitive chessParseability
Structure & completeness
Impact & quantification
Skill coverage
Timeline integrity
Language hygiene
No applicant tracking system gives candidates a score. This is the Revquix Readiness Score - ours, published, and versioned. Anyone claiming to reproduce a real ATS score is guessing.
Weights for a mid-level resume - the stage yours was scored at, read from your own dates.
Rubric version r1. Every report records the rubric it was scored under, so a score you took a screenshot of stays explainable even after we publish a new one.
End of the sample report. Everything below this line is the page's own copy - how the score is built, what we do with your file, and the questions people ask before uploading one.
Run it on mineOnly want to know whether a parser can read your file? That question is free and needs no account - the ATS checker makes no external call at all, so it costs us nothing to run.
Check parsing, freeYour text is redacted before it reaches any AI model: name, email, phone, address and date of birth are replaced with tokens first. The model reads your document and writes the explanations; it never chooses a number - every score is arithmetic over a fixed rubric in our own code. One input to that arithmetic is the model's reading of which skills you evidence, so we store your result and re-running the same input returns it rather than a fresh one. We never invent a figure your own material does not contain.
Credits are a licence to use these tools - not money, not transferable. Full detail in our privacy policy and AI policy.
What it checks, what it deliberately does not, and how to read the output.
The Revquix Readiness Score is arithmetic, not a judgement call. Eight dimensions are scored 0–100 independently, then weighted and summed: parseability, structure and completeness, impact and quantification, skill coverage, timeline integrity, language hygiene, career trajectory and scope, and screening risk. The report shows each dimension's sub-score, the weight actually applied, and the exact number of points it contributed - and those contributions add up to the total in front of you.
The weights are calibrated to career stage, because a fresher and a director are not the same document. A fresher is weighted towards demonstrated skills and complete sections; a leader towards trajectory and the scale of what they have owned. We tell you which stage we read you as, and what in your own dates produced it, so you can argue with it. Every stage's full weight table is published - not only yours.
Bands are fixed thresholds rather than model labels: 0–49 needs work, 50–69 getting there, 70–84 strong, 85–100 excellent. Because the band is deterministic, the same file scores the same on every run for a given rubric version - which matters more than it sounds, because people screenshot these numbers and compare them with friends.
The language model in this pipeline is handed the findings our own code has already computed and asked to explain them and rewrite your bullets. It is not asked to score anything, and it structurally cannot: the JSON schema its response is validated against contains no numeric fields at all.
This is the difference between a score you can argue with and a score you can only accept. Most resume scorers let the model do the scoring, which is why running the same resume twice through them gives two different numbers. Ours is computed in Java against weights you can read.
It will not invent your facts. Where a bullet needs a number and your resume does not contain one, the rewrite inserts a clearly marked placeholder and lists it separately, flagged in the interface - it never asserts a metric you did not supply. A fabricated figure you paste into a real resume is an interview you fail.
It is also not an ATS score. No applicant tracking system emits a candidate-facing score, so anyone advertising one is describing something that does not exist. This is our own published rubric, and it says so on every report.
The ones that come up before somebody uploads anything.
No. No applicant tracking system produces a candidate-facing score, so any product claiming to show you your "ATS score" is showing you something it invented. This is the Revquix Readiness Score: our own rubric, published in full and versioned, with every deduction traceable to a weight you can read.
Yes, for a given rubric version. The arithmetic is deterministic - no dependence on the current date, and none on the order findings were assembled in. Where a model reads your resume to decide which skills it demonstrates, that reading is cached against the document itself, so re-running the same file replays the same verdicts rather than asking again. If we change the weights we publish a new rubric version, and every report records the version it used.
No, and this is enforced rather than requested. Rewrites restate your own material. Where a bullet clearly needs a figure your resume does not give, the output inserts a marked placeholder and lists it in a separate placeholders section so you can fill it in with a real number.
Two credits per full report. A verified signup grants 20 credits, so the complete report is free ten times over before you pay anything, and there is one free deterministic run available every day regardless of balance.
It is stored to produce the report and then deleted - within 24 hours for an anonymous run, within 30 days for a signed-in one, unless you explicitly save it to your profile. Before any AI call, your name, email, phone number, address and date of birth are replaced with tokens.
No, but it makes the skill-coverage dimension considerably more useful. With a target role we can list the skills that role expects and which of them your resume does not evidence. Without one, that dimension scores evidence and density only - it degrades, it never errors.
More from Resume & Profile, or browse the full catalogue.
Every automated check has a ceiling: it can tell you a bullet has no number in it, but it cannot decide which of your projects is worth two lines and which is worth six. A Revquix resume review is written by a person who hires for your target role.
Resume Review from ₹299