Resume matching using AI

In this case, we used the data from the client’s ATS to create an AI that accelerates the matching process.


Our customer is a recruitment company based in France. They previously relied on manual processes for matching candidates with job postings, which was time-consuming and generated a large amount of data.

The goal of the project was to use this data to develop an AI-powered solution that would automate the matching process and provide more efficient and accurate results.



During the course of this project, we encountered several challenges related to natural language processing. These challenges included the use of different languages, variations in the length and structure of resumes, and other factors that made it difficult to accurately and consistently process the data.

We also faced challenges in integrating the data into our system, as it was stored in an external application tracking system and needed to be accessed via API.

In addition, we had to carefully plan our (re)training strategy to ensure that the AI model was able to handle the large volume of resumes and job postings in a timely and efficient manner. Overall, these challenges required us to be creative and resourceful in order to successfully deliver the project to our customer.


For this project, we built an ETL (extract, transform, load) pipeline to extract data from the Talents’In API, integrate it into our local systems, and train the AI model on a daily basis. The AI model was precomputed to optimize query-time performance, which allowed it to quickly and efficiently suggest the best candidates for a given job posting.

As a result of this project, our customer was able to significantly reduce the time and effort required to source candidates, thanks to the AI-powered matching system we developed.