Importing a CV
Starting from a PDF, Word file or image, starting from JSON Resume, and what happens to the file you upload.
Start a new CV from a file.
Two routes in, and they work very differently.
From a document (PDF, Word or an image)
Upload a CV file takes a PDF, a Word document, or a photograph of a CV, and extracts the content into the editor.
The file is sent to OpenAI to extract the data. Steerd tells you this before you upload: "Your file is sent to OpenAI to extract the data. It is not stored."
That disclosure is there so the decision is yours. If the CV belongs to somebody else, they should know; if it carries information you are not willing to send to a third party, use the JSON Resume route or type it in.
Extraction is good, not perfect. Read what it produced before you send it to anybody. Dates and job titles usually come through cleanly; nuance in a summary sometimes does not.
From JSON Resume
Upload JSONResume takes a .json file in the open
JSON Resume format. No third party is involved: the file is parsed
directly, because the format is already structured.
If the file is not valid JSON Resume, Steerd says so specifically rather than failing vaguely.
After importing
You land in the editor with the content filled in and nothing saved yet. Fix what the import got wrong, then create the CV.
The imported CV is named for you and can be renamed.
Which to use
If you already have JSON Resume, use it. If you have a PDF and are comfortable with the extraction step, that is by far the fastest way to get a decade of history into a structured editor.