Resume Scoring Algorithm

Project Description

We source and hire thousands of independent contractors for our clients. We have a rigorous test-based application process that all candidates go through. Given the data we have collected after vetting over one hundred thousand applicants, we'd like to start finding correlations between resume/linked-in profile attributes and test performance. The goal is that we would be experts at identifying the candidates that are most likely to do well on our tests.

In order to conduct a proof of concept, we'd like to start by providing you with data for three of our Pipelines: Java Software Engineer, Java Chief Software Architect, and .Net Chief Software Architect. We have the test data and resumes for these candidates (14,000+) in AWS. We may also have some team members manually add their linked-in profiles to increase the amount of data you have to analyze. As far as the actual format of the data, we'd like to discuss the best way for you to get it. You can talk to our engineering team to finalize those details. Regarding output, we would want you to identify the top indicators of a successful Java Software Engineer candidate. We'd like for you to consider a multitude of factors (skills, job history, location, work experience, among others). We can ennumerate that list of factors along with you when we discuss this in more detail.

If we prove the concept of this theory we'd definitely pursue the same analysis for many more pipelines and ultimately try to implement a closed-loop system that allows us to learn from the performance of all of our candidates. 

Here is a small sample of what the data we give you would look like: https://docs.google.com/spreadsheets/d/18Xvt4ib9mTrxgrcNWpJQboQq8HMi6Snn4IMGhJy6xlk/edit#gid=1291488262