Your CV and facts
How jlog reads your CV into facts, and why you review them before anything is stored.
Why jlog reads your CV
Two features need it. Autofill uses your name, contact details and links. Tailored CVs and drafted answers are built only from the facts in it, so they can never claim something you did not write.
Importing
Open CV in the sidebar and import one of:
- A PDF. Its text is read in your browser. A scanned PDF has no text to read; paste the text instead.
- Pasted text, including Markdown or LaTeX. The format is detected for you.
Plain text from a PDF is structured by your AI provider when you have one set, and by jlog’s own rules when you do not.
Reviewing the facts
Before anything is stored, jlog shows every fact it read, one claim per line, under the role it belongs to. Each one can be traced back to a line of your CV.
- Tick what is true. What you leave unticked is not stored and never used.
- Reworded lines are dropped. A fact must be copied from your CV, not paraphrased. If the model rewrote a line, it is set aside and listed, so nothing new slips in.
- Lines with no role above them are listed separately rather than guessed into one.
Your profile
The same page holds your name, contact details, links and an optional photo. Autofill fills forms from these, and a tailored CV puts them in its header.
The original file
When you import a PDF, the file is kept so a tailored CV can show each line over the page it came from.