Mosaic

• 

  Applicants listed with pseudo-names for screening

• 

  Bias inducing information hidden for screening

• 

  Resume with bias inducing information redacted

• 

  Job description bias checker

Inspiration

We decided to conduct research and found a plethora of information on current hiring softwares. However, a lot of these softwares included the human touch in some way. Knowing that the main problem was the human bias, we thought about the ways that we could remove the human bias. First we thought maybe we could automate the whole process, interview includes, using AI, but that proved to be too daunting a task. We then took inspiration from one of the studies we found, where researchers of a study replaced names of applications with gender typical names, with all else being equal. We thought that instead of automating everything, we could just replace or hide the potential differentiating data.

What it does

In order to reduce gender bias, Mosaic provides a blind recruiting process and job description checker.

Blind Recruiting Process

During the blind recruiting process, Mosaic first collects data through applications.

Then, it hides or alters details that could affect the recruiters' choice. For example, names will be changed to (e.g. White Peacock), and any pronouns in their resume would be changed to gender neutral pronouns. Furthermore, other data such as the applicant's school, contacts, birthday, gender and referral sources will be hidden as they may induce bias during the hiring process.

When the recruiter has scheduled an interview with the applicant, the applicant's full details will then be revealed.

Job Description Checker

Employers can also use Mosaic to check if they have unintentionally introduced gender bias into their job descriptions.

To do so, Mosaic will use parse the job description and check for any gendered terms based on research.

Mosaic will then highlight any gendered terms and offer gender neutral alternatives for them.

How we built it

We built Mosaic as a pure HTML, Javascript project to minimise the realisation time. However, we intend to use Vue.js and Firebase when refining it in the future.

Challenges we ran into

The main challenge we ran into was the lack of time. Our team had difficulties coming up with ideas and settling on one. As such, we spent a majority of our time researching into gender biases during recruiting and possible areas to target. Due to the lack of time, we had to use a simple tech stack that required little time to set up. As such, the prototype is not fully functional yet. However, we strongly believe in our project and what it stands for, and will continue to improve Mosaic.

What we learned

We have learned more about the problem of bias in hiring, and how blind interviewing and gender neutral job descriptions can help improve diversity and inclusion. With this, we hope that employers will be able to hire women who have less social mobility.

What's next for KAPY

Apart from making Mosaic fully functional, we would like to also expand Mosaic to:

1. Have an automated chatbot that can handle initial recruitment inquiries, such as whether the applicant possesses certain certifications or qualifications.

2. Track the applicant throughout the different hiring stages.

3. Filter applications to ensure diversity at each hiring stage.

Built With

• css
• html
• javascript