How the CV is parsed ?
We take the CV document and try to find different parts :
- Professional experiences, trainings and educational trainings with title, description, begining date, end date, location and company
The CV is splitted in three parts to find the professional experiences, the professional trainings and the educational trainings parts. Each type of experience is then formatted to find the title, the description, the beginning and end date, location and company.
- Skills (as stated in the CV)
This is different from skills extraction where we try to infer skills from a text. Skills extraction is also done on a CV with live extraction. Here we are talking about finding a skill section on the CV and parsing it.
We also generate a short summary from all those info about the person. We use LLM prompt for each part.
Whatever the langage of the CV is, we transform it to the langage set in the application by the user.