ExtractFoxExtract data from resumes and CVs
Turn resume PDFs and scanned CVs into structured candidate data: contact info, work history, education, and skills — ready to import into an ATS or compare across a stack of applicants.
Why this matters
Resume parsing has traditionally needed brittle rule-based parsers per layout. With a multimodal LLM, ExtractFox handles wildly different CV designs — including infographic-style layouts and two-column templates — without per-template tuning.
How it works
- Step 1Upload the resume
PDF or image. Multi-page CVs are fine.
- Step 2Get structured profile data
Name, contact, headline, summary, skills, and full work and education history.
- Step 3Send to your ATS
JSON for direct API ingest, or Excel for a candidate spreadsheet.
Fields extracted
full_nameemailphonelocationheadlinesummaryskills[]experience[].companyexperience[].titleexperience[].start_dateexperience[].end_dateexperience[].descriptioneducation[].institutioneducation[].degreeeducation[].fieldeducation[].start_dateeducation[].end_dateSample output
Example output from a software engineer resume
| full_name | Jane Doe |
| jane@example.com | |
| phone | +1 415-555-0182 |
| location | San Francisco, CA |
| headline | Senior Software Engineer |
- TypeScript
- React
- Node.js
- PostgreSQL
- AWS
| company | title | start_date | end_date | description |
|---|---|---|---|---|
| Acme Corp. | Senior Engineer | 2023-06 | — | Led migration to event-driven architecture; mentored 4 junior engineers. |
| Beta Industries | Software Engineer | 2020-01 | 2023-05 | Built customer-facing analytics dashboards in React. |
| institution | degree | field | start_date | end_date |
|---|---|---|---|---|
| UC Berkeley | B.S. | Computer Science | 2016-09 | 2020-05 |
Frequently asked questions
How do I parse a resume PDF into structured data?+
Drop the PDF here, click Extract, and you'll get a JSON object with name, contact, headline, skills, full work history, and education. Download as JSON for an ATS import or as Excel for a recruiter spreadsheet.
Does it work on infographic-style or two-column resumes?+
Yes. The model reads visual layout, not just text flow, so designer CVs and two-column templates parse as well as plain ones.
Can I integrate this with my ATS?+
Yes — the JSON output is stable and easy to map into Greenhouse, Lever, Workable, or a custom ATS. The REST API is on the paid plan.
Does it extract dates and durations consistently?+
Dates come back in YYYY-MM format where the resume specifies a month, or YYYY when only the year is shown. Current roles return null for end_date so you can compute tenure.
Can I extract data from a LinkedIn profile?+
Yes — see the dedicated LinkedIn profile extractor. Save the LinkedIn profile to PDF (Profile → More → Save to PDF), drop it there, and you'll get a structured candidate object including headline, current company, and tenure that the resume schema doesn't surface.
What about non-English resumes?+
Multi-language resumes work. Names, companies, and dates extract reliably; long-form descriptions come back in the original language.