We started off as a recruiter hack that would go through GitHub and Stack Overflow data based on the information given in the candidate's resume. As we developed, we realized that we could expand the scope of the project. We came up with a scoring system RICO Score changes the traditional dynamic with recruiters giving candidates more power of jobs. It will enable companies to identify qualified and ambitious candidates that fit the job description.

What it does

The web application is targeted for company use. The recruiter would insert a job description and upload resumes of candidates that have applied for the job or are being considered for the job to find the RICO score for each one to determine their fit for the position.

The iOS app is targeted for job seekers. Job seekers create a profile by logging in with their GitHub account, and by leveraging their social presence, candidates can find jobs and connect with companies that match their qualifications and skill level. Job seekers will be given a RICO score based on the information we get access to.

The RICO Score is calculated under the following scoring system:

Match 1 - Competency based on social level

0/50 - Resume competency score

0/10 - Stack Overflow score

0/40 - GitHub score

Match 2 - Match on Job Description

0/100 - Percentage of tags of job description that match with a user's tags

How we built it

The backend is written entirely in Python, the front end is mainly html/css with a hint of JavaScript. We have also created an iOS app, which was inspired by Tinder, written in Swift - we have used a firebase backend for the iOS app.

Challenges we ran into

We ran into a wide range of challenges in different stages of development:

  1. Coming up with a scoring system was one of the major challenges we faced, we improved on and made changes to the scoring system as we developed.
  2. We ran into issues with the APIs we used, and we kept surpassing the request limit which restricted our testing capacity:

-- GitHub doesn't have a user's total commits anywhere so we had to calculate them manually. This was much more complicated than it seems as one must iterate through every pull request made, every repository created, and every repository the user has contributed too and iterate through all those commits to find commits by the target.

-- Stack Overflow API is severely limited in how one can pull user data. So to get around this, we scan the dataset provided by Stack Overflow for users with the same names as our target, and we do geolocation checks against the results until we narrows it down to one user in the area. From there we glean their reputation and use that to calculate our RICO Score along with the GitHub and resume competency score.

Accomplishments that we're proud of

  1. The scoring system - it is a modulus scoring system where you can take things out and put it back in easily.
  2. Creating a web and iOS app in less than 24hrs.
  3. A couple of us in the team had not developed in Python before, so we are proud of learning something new and creating something with it.

What we learned

  1. How to handle complicated json data in python
  2. Learning how to implement ajax post request
Share this project: