This is a small pet project that helps me with the building of my CV and deploy it to my website in an automated fashion.
It reads from a single source of truth (a JSON file), and through a Jinja2 template, it generates a .tex file, which is then compiled into the final PDF file.
For the deployment, another repo is used. This repo is private because my personal data is there. If you have a real need for it, drop me an inbox and I can analyze your request and grant access.
Throughout my career, I have had to write many versions of my CV. As a programmer, I do not like to repeat myself, so copy-pasting data from one CV standard to another seemed to be a little bit of a re-work. Initially, my idea was to use LinkedIn as a source of truth for my data, but they don't have a good API for that. Therefore, I decide to take control of my own data and make this little project.
Another big reason for doing this project is my believe that you should own your on data, so here we are :)
A sample PDF compiled used this project can be found here.
If for some odd reason you want to run this own your own, it should be straight-forward. There is a Makefile that does most of the heavy-lifting.
I want to have various CVs that I can tailor to a specific need or market. For example, in Switzerland, CVs with photos are well-received—on the other hand, in Brazil, this is frowned upon. Adjusting a CV for a specific role —for example, by changing some wording or emphasizing some skill set—is also something that I want to look further. Also, exploring LLMs to rephrase some wording for some specific context might sound like a good idea.
This project is licensed under the MIT License.