CVs in any language
Send CVs in any language. Free text (job titles, descriptions, quotes in
evidence) stays in the CV's own language, and the values your code branches on, such as
enums, ISO dates and country codes, are always the same English or ISO forms.
What that means in practice
- Headings: sections are classified by content, so "Berufserfahrung", "Experiência" or "Where I've been" all work.
- Dates: "fev. 2022", "seit 2021" and "03.2020" are parsed by code into ISO values with a precision.
- Raw and normalised side by side: a location such as "Lisboa, PT" becomes
{"city": "Lisbon", "country_code": "PT"}with the raw text kept. - Names: never transliterated; they come back as written.
- Mixed-language CVs (for example Portuguese and English) are handled in one pass.
- Scans: OCR picks languages by script: Latin, Cyrillic, Greek, Arabic, Hebrew, Devanagari, Chinese, Japanese and Korean.
spoken_languages lists the languages the candidate says they speak, with the level
they give (for example C1 or native).
Try it
Get a free API key (50 parses a month, no card), then:
curl https://api.verbatimcv.com/v1/parse \
-H "Authorization: Bearer $VERBATIMCV_API_KEY" \
-F file=@cv.pdf
Agents can use the parse_resume tool instead: uvx verbatimcv-mcp. See the
quickstart for Python, JavaScript and MCP setup.