Total years of experience
Read derived.total_experience_months. It is calculated in code, not
by a model: the sum of every role's date range with overlaps counted once. The method is returned
next to the number, so you know exactly what it means.
"derived": {
"total_experience_months": {
"value": 98,
"method": "sum of non-overlapping ranges; year-only dates counted from January"
},
"seniority": { "value": "senior", "method": "title keywords + experience length" }
}
Edge cases handled
- Overlapping roles (a side contract during a full-time job) count once.
- Year-only dates ("2018 – 2020") count from January, with a
DATE_YEAR_ONLYwarning on the date. - Current roles ("Present", "Presente") count up to today.
- Imprecise totals can appear in
agent_guidance.confirm_before_acting, so an agent doesn't filter someone out on a rounding error.
Every date keeps its raw text, so you can show the source next to the total.
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.