A CV machines can actually read.

It's an open standard for publishing your CV as a single JSON document at a stable URL, like livelink.cv/ashley/cv.json. People see a beautiful page. Machines (AI recruiters, ATS, agents) read the structured data directly from the same source. One file, two audiences, always in sync.

A PDF is a frozen picture of text. Machines have to OCR it and guess at the layout, which is exactly where good candidates get silently dropped. cv.json hands them clean, named fields instead, like "skills", "availability" and "meta", so nothing is misread. And because it's one live link, you never re-send an outdated copy.

We love JSON Resume and stay field-compatible where it makes sense, so migrating is painless. cv.json adds the parts hiring actually needs in 2026: discovery (so agents find your file automatically), versioning (an X-CV-Version header + meta.version), hiring signals (open-to-work, roles, work type, visa) and privacy by default (contact hidden unless you choose otherwise).

Completely. The spec is MIT-licensed and lives on GitHub, so fork it, host your file anywhere, and move it any time. You can create your own cv.json for free with the FreeCV builder, and your data is always portable and yours. No lock-in, ever.

It depends on the system. Most legacy ATS still expect a PDF or Word upload and extract the text from it. Modern AI recruiters and agents increasingly fetch structured data instead, and cv.json is built to be read directly over open CORS. Publishing a cv.json means the machines that can read structured data get clean fields, while you still hand a normal PDF to the ones that can't yet.

A machine-readable resume is career data stored as structured, named fields that software can read without guessing, rather than a picture of text a parser has to reverse-engineer. cv.json is a machine-readable resume: every skill, role and date sits in a labelled field at a stable URL that AI recruiters and ATS can fetch directly.

AI screening tools reward clean structure and real, readable text. A single-column layout with standard section headings parses far better than multi-column designs, tables, or text trapped inside images. A machine-readable source like cv.json removes the guesswork entirely, because the data is already structured before any tool reads it.

Resumes are read by software before a human ever sees them, and that only deepens as AI agents start doing first-pass screening. The format is moving from a static file you attach to a live, structured profile that systems fetch and trust. cv.json is built for exactly that future: one URL, always current, readable by people and machines alike.