Problem Statement
Availability of skilled resources is always a challenge for any industry. More so for the IT industry. Potential Hires to Job Description matching was done manually in old days. With the advancement of Machine Learning and particularly Natural Language Processing algorithms, the task of recruiters have become easier. These techniques probably have helped recruiters to search the proverbial needle in haystack.

Problem Dimensions

This problem has the following dimension

  • Parse resume in varying formats (.docx, .doc, .pdf, .txt, etc) to store in a database
  • Build a Parser such that it should be able to extract ROLES, SKILLS, PERIOD, EDUCATION, COMPANIES, besides the usual attributes of NAME, EMAIL, PHONE number and store it in database
  • The Parser should also be able to build/ synthesize/ extract a synopsis of the Candidate’s profile.
  • Solutions will be judged on the ability to extract accurate information in the above fields in the database for different formats and filetypes of resume

So That
The recruiter is presented an easy to use GUI to initiate contact with candidates.
For e.g
July 2001 – Aug 2003 – Software Engineer – worked on Core Java on such and such project
Sept 2003 – Mar 2010 – Senior Software Engineer – Version control worked on TOAD and GIT
Apr 2010 – Jan 2015 – Project Lead – managing a project on Hadoop and Dockers