Free match check — a basic check, then a prompt for your own AI
Upload. Paste the job. Check.
You'll get a match score and a smart prompt to paste into ChatGPT, Claude or Gemini. There's no right answer to screening — this just gets the basics right.
What this actually measures
Most “ATS scores” are a model’s opinion dressed as a number, which is why running the same resume twice gives you two answers. This one is arithmetic. A model reads the posting once and extracts the terms it asks for — separated into required and nice-to-have, with aliases so “K8s” and “Kubernetes” count as the same thing. Your resume is then checked against that list by a fixed formula, the same one the app uses. Under the score you get four format notes — numbers in your results, bullets that open with what you did, a readable length, a way to reach you — which are rules too, and do not move the number.
That has a real limit and it is worth stating plainly: coverage of the words a posting uses is not the same as being the right candidate, and no checker can tell you the second thing. What it does tell you is whether a recruiter scanning for the six things they were told to look for will find them — which is a solvable problem, and usually solvable without changing a single true fact about your experience.
A missing term is not an instruction to add it. If the posting wants Kubernetes and you have never touched it, the honest answer is that this is not your job. The gaps worth closing are the ones where you did the work and your resume calls it something else. The prompt tells your AI the same thing: it may not invent experience, and it has to list the terms it could not honestly add.
Then what?
Copy the prompt into ChatGPT, Claude or Gemini and it comes back with each edit as before → after, with a reason tied to the posting. Pick a full tailor, the five highest-impact edits, or a short cover note.
Or skip the pasting. With a free account, the same posting produces proposed changes on your document — each one a tracked change carrying its reason and what it does to the score. You approve them one at a time, or decline them. Nothing is applied on its own, your original resume is never overwritten, and you export the finished version as PDF, Word or plain text.
5 tailored resumes a month, free. No card.
Questions about the checker
- Is this really free, and do I need an account?
- The score, every missing term and the prompt are free and need no account. What needs an account is having the changes made for you — proposed on your document as tracked changes you approve or decline one at a time — because that is the part that costs us a model call.
- How is the score calculated?
- A model reads the posting and extracts the terms it asks for, weighted by whether they are required or nice to have. Your resume is then scored against that list by a fixed formula on the server — the model never produces the number. That is why asking twice gives the same answer. The notes under the score (numbers, action verbs, length, contact) are separate rules and do not change it.
- Is it the same score the app gives?
- Yes. It is the same term extraction and the same formula the app uses. Sign up from the result and the resume you checked comes with you, so there is nothing to upload twice.
- Does a high score guarantee I get through the ATS?
- No, and anyone claiming otherwise is guessing. Applicant tracking systems differ, most rank rather than reject outright, and a human still reads the shortlist. What this measures is concrete: how much of what the posting explicitly asks for actually appears in your resume. That is worth fixing regardless of which system is on the other end.
- What happens to the resume I upload?
- The file is read once and deleted — we keep the text for a day at most so it can be scored, never the file. If you create an account from the result, that text becomes your first resume. The prompt contains your resume text, so when you paste it into ChatGPT, Claude or Gemini, that service's own terms apply.
- Why does it say my PDF cannot be read?
- Almost always because it is a scan — an image of a document rather than a document. There is no text in it to extract. Export a fresh PDF from the original file (Word, Google Docs, Pages) and it will read fine.
How to tailor a resume to a job description · What an ATS actually does · ATS-safe templates