Parse any CV. Even the ones nothing else can read.
Scanned PDFs. Phone photos. WhatsApp CVs. Spanish, Polish, German resumes. Recruitly reads every one of them and extracts every field — no manual typing, no per-CV fees.
Scanned PDF
image-only, 4 pages
WhatsApp photo
IMG_2049.jpg
Phone scan
rotated 90°, skewed
Spanish CV
scanned, accented characters
Candidate record
Every CV format your candidates send
The files your current ATS rejects, Recruitly reads. No special workflow. No re-typing. Same parsing pipeline, every file.
Scanned PDFs
Image-only PDFs from a printer, copier, or scanner. No selectable text, no problem. Recruitly detects the missing text layer and runs OCR automatically.
Phone-photo CVs
A candidate snapped their CV on a phone. JPG, PNG, HEIC — Recruitly corrects rotation, fixes skew, and extracts every field.
WhatsApp CVs
Recruiters everywhere get CVs over WhatsApp. UK construction, hospitality, healthcare, social work — every desk sees it. Drop the file into Recruitly — it parses, no manual typing.
Every European language
Spanish, Polish, German, French, Italian, Portuguese, Romanian, Bulgarian, Dutch — every CV the UK and EU sees. Arabic, Hindi, and CJK on top for international desks.
Cloud + local fallback
Primary OCR runs on Google Gemini for accuracy. If that ever has an outage, local PaddleOCR and Tesseract engines take over. Pipeline never goes down.
Every field, extracted
Name, email, phone, location, current role, full work history, education, skills, languages. Same field set as a native PDF parse — populated straight into the candidate record.
Why other parsers fail on image PDFs
Most ATS parsers only look for embedded text. If a CV is a scan or a photo, there’s no text to find — so the file is silently rejected or imported as a blank record. Recruitly runs a four-stage pipeline that handles every format.
Detect whether the file has embedded text. If yes, parse directly. If no, route to OCR.
Try cheap, fast local engines first (PaddleOCR, Tesseract) for clean scans.
Escalate hard pages — photos, multilingual, low contrast — to Google Gemini for accuracy.
Extract structured fields into the candidate record. Low-confidence fields flagged for review.
Embedded text?
PDF, DOCX with text layer
Local OCR
PaddleOCR · Tesseract
Cloud OCR
Google Gemini 3.1 Flash Lite
Field extraction
Name · contact · history · skills
The CVs other ATS systems quietly drop
Every rejected CV is a candidate you never speak to. Recruitly imports the ones your competitors lose.
| File type | Most ATS parsers | Recruitly |
|---|---|---|
| Native PDF / Word CV | ||
| Scanned image-only PDF | Often fails | |
| Phone-photo CV (JPG / HEIC) | ||
| WhatsApp CV | ||
| Spanish / Polish / German CV | Limited | |
| Rotated or skewed scan | Often fails | |
| Outage-resilient OCR | ||
| Per-CV OCR fee | Common | $0 |
Built into the platform. No per-CV charges. No OCR add-on.
Everything the parser handles
One pipeline, every input format your candidates send.
Stop losing candidates to a parser.
Try Recruitly on the CVs your current ATS keeps rejecting. We’ll read them.