Method
How this works
Most resume scanners score your wording against a job description and call the result an "ATS score". This one does something narrower and more verifiable: it reconstructs the text a parser would extract from your file, and tells you where that differs from what you wrote.
1 — Your file is read in this browser
The PDF is parsed by pdf.js, the same engine Firefox uses to display PDFs. It runs as JavaScript in your tab. There is no upload, no server, and nothing stored — you can disconnect from the network after this page loads and the check still runs.
2 — Position is kept, not just text
A PDF stores instructions to draw characters at coordinates. Reading it as plain text throws away the coordinates, and the coordinates are where the problems live. Every text run is kept with its x and y position, its font, and its size, then grouped back into visual lines by baseline.
3 — Reading order is reconstructed twice
Once the way a human reads it — down each column in turn — and once the way a line-based parser reads it, straight across the page. Columns are found by projecting every text run onto the horizontal axis and looking for a whitespace gutter that runs most of the page height with body text on both sides of it.
When those two readings disagree, the resume is being misread, and the report shows you both so you can judge the damage yourself.
4 — Twelve rules are applied
Each rule is deterministic. The same file always produces the same result, and every finding quotes the evidence from your own document that triggered it.
- TEXT_LAYER
Resume contains selectable text
- COLUMNS
Single-column layout
- CONTACT
Contact details are machine-readable
- TABLES
No table-based layout
- HEADER_FOOTER
Nothing critical in headers or footers
- HEADINGS
Standard section headings
- ENCODING
Text encodes to real characters
- DATES
Consistent, parseable dates
- GRAPHICS
No information locked in graphics
- FONTS
Conventional, readable typography
- LENGTH
Sensible length
- FILENAME
Professional filename
What this does not tell you
- Whether you match a specific job. Keyword fit is a different question and depends on a posting this tool never sees.
- What one particular system does. Workday, Greenhouse, Taleo, and iCIMS each parse differently, and none publish their internals. The checks here target failures that are damaging across all of them.
- Whether your resume is good. A perfect score means a machine can read it. Whether it is worth reading is still up to you.
On "ATS scores" generally
Treat any tool that gives you a percentage match against an employer's system with suspicion. No vendor exposes that number. What can be measured honestly is whether the text survives the trip — which is what the score here means, and nothing more.