When you upload a CV to a job portal, a parser reads it in a few steps. First it gets the text out of the file — directly from a text-based PDF or Word document, or through OCR if the file is a scan or photo. Then it works out the structure: which lines are headings, where each job starts and ends, which dates belong to which role. Finally it saves the pieces as fields — employer, title, start and end dates, school, skills — that an ATS can filter and search. Modern parsers often use AI models for the middle step, which helps with unusual wording but still depends on a readable file.
Resume Parsing (CV Parsing)
Resume parsing is the automatic extraction of information from a CV file — name, contact details, job titles, dates, education and skills — into structured fields that software can store and search.
Try it now — no sign-up needed
What it means
From file to fields
Why it matters
Misread CVs go missing
If a parser puts your job title in the company field or drops a role entirely, recruiters searching the database may never see that experience. Common causes are multi-column layouts read in the wrong order, text inside images or text boxes, contact details in the page header, and dates written in mixed formats. Many portals let you review what they extracted before you submit — always check it and correct any field that came out wrong.
How Twin CV helps
Parsing that works for you
Twin CV uses parsing in your favour: CV import reads a PDF, DOCX, TXT or photo of your old CV and fills your profile, experience, education, skills, languages, certificates, projects and extras. There's no review step before saving, so check the result and fix anything off — by hand or by telling the assistant. Dates come in as years, and your photo isn't imported.
FAQ
Questions about CV parsing
Do parsers read PDFs?
Most do, as long as the PDF contains real text. A scanned PDF is just an image, so the parser needs OCR to read it — and results vary.
How do I check how my CV parses?
Upload it to a portal that shows its extracted fields, or copy all the text from your PDF into a plain text editor: if the order looks jumbled there, a parser may struggle too.