My position was terminated.
So now I am looking for a new one.
I already had a CV. Cord.co generated it, and for a while that was enough. It looked tidy. It had the right sections. I could send it to someone and not feel embarrassed.
Then I looked at it again, the way a stranger would.
There was a wordmark that was not mine. A footer that was not mine. Colors that belonged to the template, not to the page I actually wanted to hand over.
It is a small thing, until it is the document that is supposed to represent you.
I did not want a third party sitting at the bottom of my own work history.
What I actually wanted
One file with the content.
One page that renders it.
Print that looks like paper, not like a website squeezed onto A4.
No account. No export button that still leaves their brand on the result.
cv.json is the CV. Everything else is just a way to look at it.
The page
It is a static page. It fetches the JSON and draws the sheet: photo, name, summary, then two columns. Contact, skills, preferences and languages on the left. Experience, education and projects on the right.
The first version was only that. Open it, read it, print it.
That already solved the branding problem. The paper is blank except for what I wrote.
Then I started using it for real, and the paper had opinions.
A section heading would sit at the bottom of a page, and the list under it would start on the next one. On page two the left column and the right column did not share a top edge, so "Languages" sat lower than the job title next to it. The contact block wanted to stretch across the whole page the moment you printed.
CSS break-inside helps until it doesn't, especially once a column is a float and the PDF is no longer the browser's problem.
So the print layout is explicit now. The left column stays a left column on every page. A block that does not fit is moved as a whole: the heading and the lines under it travel together. The first heading on the next page lines up with the first heading beside it.
A PDF you can actually select
The first PDF export was a screenshot. html2canvas, then jsPDF, one JPEG per page. It looked right and it was useless the moment someone tried to copy an email address out of it.
The download is a normal PDF now. Text is text. Links are links. The photo is the only image.
Job titles and dates share a line when they fit. When they don't, the date drops under the title, aligned right, instead of slicing the company name in half.
Why bother
I could have kept the Cord PDF and cropped the footer in a graphics editor. I have done dumber things under a deadline.
I didn't want the next five years of updates to start with "open someone else's product, export, then clean up their marks."
The CV is a JSON file in a folder I control. If I change a job, I change the file. The page and the PDF follow.
That is the whole tool. You can use it here if you want to. Enjoy!
Best,
-- ł.