Paste your CV and the job advert, and see which of the advert's terms actually appear in your CV. Keywords are pulled from the advert automatically by how often they appear, or you can list them yourself. Matching is on whole words, which matters more than it sounds: a substring check reports Java as present in a CV that only says JavaScript, and that false positive is what makes most of these tools worthless. Phrases like Spring Boot must appear as that phrase, and terms with symbols like C++ or .NET are handled properly. Your CV is never uploaded. One honest caveat — this is a word check, not an applicant tracking system.
| CV says | Looking for | Substring check | This tool |
|---|---|---|---|
| JavaScript | Java | Match | No match |
| NoSQL | SQL | Match | No match |
| Spring is nice. Boot up. | Spring Boot | Match | No match |
| Built with C++ | C++ | Often missed | Match |
No. Everything is compared in your browser. A CV is full of personal detail — name, address, employment history — which is exactly what should not be pasted into a server-side tool.
No, and treating it as one is a mistake. Real applicant tracking systems parse your document's structure, dates and job titles and weight terms by where they appear. This counts word matches. Use it to find terms you have genuinely left out, not as a pass mark.
Because matching is on whole words. A substring check would count them as the same, which is a false positive that tells you nothing useful. The same applies to SQL and NoSQL.
Yes. Word-boundary matching normally fails on terms that start or end with punctuation, so the boundary is only applied at the edges where the term itself has a word character.
Only the ones that are true. Adding skills you do not have gets you through a filter and then fails at the interview, which wastes more of your time than the application saved. Use the list to spot things you do have and forgot to mention.
Paste the text, or run the PDF through the PDF Text Extractor first and bring the result here.
Terms are matched as whole words rather than substrings, which is the difference between a useful result and a misleading one: substring matching reports Java as present in a CV that only mentions JavaScript, and SQL as present in NoSQL. Word boundaries are asserted conditionally, because the standard boundary assertion fails around terms that begin or end with punctuation — a boundary before C++ or after .NET never matches, so those terms would be reported missing however often they appear. Multi-word terms are matched as phrases, so Spring Boot requires that phrase rather than the two words appearing separately in unrelated sentences. When no terms are supplied, they are extracted from the pasted job description by frequency with stop words removed, which surfaces what the advert actually emphasises rather than what a generic list assumes. The score is simply the proportion of supplied terms found, and the page is explicit that this is a word-match check and not an applicant tracking system: real screening software parses structure, dates and titles and weights terms by context, so a high score here guarantees nothing there. It is stated plainly because the alternative encourages people to stuff in skills they do not have, which fails at the interview rather than at the filter.