Check what was extracted, then compare the original
Upload a PDF to inspect text page by page, or paste text for a limited content review. The checker identifies empty text pages, conventional email patterns, selected English or Tamil section labels, spaced-out letters, unusual glyphs and long lines. Each finding explains what was observed and where to inspect it.
Why visual comparison matters
Greenhouse documents that image uploads, separated letters, complex layouts and unclear sections can cause incomplete parsing. Compare names, dates, headings and bullets with the extracted order, especially around columns or text boxes. See Greenhouse's resume parsing guidance (reviewed 12 October 2026).
Understand the limits
This uses a local PDF text reader and transparent heuristics. It does not call an employer's parser, evaluate candidate suitability, calculate a universal ATS score or predict selection. Recognized text does not prove completeness or correct order. Pasted text cannot reveal the source document's columns, images or missing content. Heading recognition uses a limited label list; other headings and languages remain valid.
No OCR is performed. Empty text pages may be scans, blank pages or unsupported encoding. Export selectable text from the original editor, or use an appropriate OCR workflow before checking again. Preview failures do not prove that the employer cannot read the file. Verify the actual application's file rules and any populated fields after uploading.
Private, bounded processing
PDF: up to 10 MiB and 20 pages. Text: up to 100,000 characters, 5,000 nonempty lines and 50,000 PDF text fragments. PDF reading has a 45-second deadline and page rendering has a 15-second deadline with a 3-million-pixel preview cap. Cancel preserves the last review; Clear removes local results. PDFs requiring passwords need an unprotected copy you can open. No resume content is uploaded, logged or saved in browser storage.