Candidate Ranking using a Job Description using Neural Networks

Candidate Ranking using a Job Description using Neural Networks

The Challenge

It is challenging to source candidates for a given job profile especially when it comes to leadership roles for an organization. One has to align the candidate personal and leadership skills to see whether he/she is a right fit for the organization. We use linkedin profile data in extracting meaningful information so that recruiters can contact the right candidates and reduce the closing time. Tatras was tasked with automating the matching process by a leading recruitment firm in the US for CXO level appointments.

Hypothesis

  • Web scraping to retrieve relevant candidate data to go beyond normalised information available.
  • LLM can assist human in filtering and getting to the desired set of candidates faster.

Execution

  • Uses Selenium to log and scrape LinkedIn.
  • Extract what we are looking for in description in terms of a mandatory skill and ideal skill.
  • Use LLM’s to extract relevant tags and give colored summary highlighting the matching and non-matching information.

Outcomes

  • 98% REDUCTION IN SEARCH TIME AND MANUAL EFFORT.
  • Uses Selenium to log and scrape LinkedIn.
  • Extract what we are looking for in description in terms of a mandatory skill and ideal skill.
  • Use LLM’s to extract relevant tags and give colored summary highlighting the matching and non-matching information.

Project Highlights

98%

reduction in seach time and
manual human efforts.