How to Build a Private ChatGPT Using Open-Source Technology? Download our free white paper.

Identify skills in a resume and cover letter using AI

Lettria can automate the reading of candidates' CVs and cover letters by using keyword detection on the applications and related to the desired profile.

How to easily identify the most relevant candidates

Many companies are facing an important challenge related to recruitment. Indeed, companies are looking for profiles that can add value to their business, but they are often faced with a very large number of applicants for each position offered.

It is necessary to be able to conduct an initial sorting of the different applications according to the academic background of the candidate, their experience and their previous positions. Generally only a minority of CVs correspond to the required profile.

Recruiters are compelled to spend time on each profile in order to qualify it for the next step, which prevents them from focusing on their core business and leads them to miss potentially interesting profiles.

Want to learn how to build a private ChatGPT using open-source technology?

Automate the pre-selection of profiles

Lettria offers to automate the reading of candidates' CVs and cover letters by using keyword detection present on the applications and related to the desired profile, which allows an initial analysis in relation to the desired skills. This first level of analysis allows us to detect very detailed information, specific to each industry and profession. To provide the best level of detail, the Lettria API is associated with a database of 750,000 words that includes numerous references to the job market (computer languages, diplomas, schools). It is also possible for each user to customize these dictionaries by adding specific notions (requested certifications, addition of specialized schools, etc.).

In addition, the Lettria APIs will be able to compare the experience of each candidate to the type of experience sought for the position by going to the cumulative experience of the candidates, according to their date of graduation and years of experience.

Finally, a matching score can be assigned to each candidate in order to propose to the HR teams the profiles that are most likely to correspond to their search. This way, teams can be sure that all profiles are reviewed and save valuable time that can be dedicated to higher value-added tasks.


Build your NLP pipeline for free
Get started ->