Inspiration
Studies show that applicants with African-American names are less likely to get an interview compared to their white counterparts. Zip codes can also be a proxy for race, which also lead to discrimination in the recruitment process.
What it does
Equalize anonymizes any personal details that could work against an applicant.
How we built it
- Python : detecting and anonymizing personal details in documents like resumes, papers and applications.
- Reactjs: web app
- Google Cloud : serverless functions
- MongoDB : data storage
Name
This is done by using a names database and a standard dictionary to analyse which parts of the file contain names. Then we compare this to a list of well known ethnically white names and replace if necessary
Address
We detect addresees in the text and lookup the statistics of the area. if below the median range of parameters, we replace some parts with a set of address details which have higher socioeconmic details
Challenges we ran into
- Extracting features from the document
- Integrating everything
- Time
Accomplishments that we're proud of
Building a functional product!
What's next for Equalize
Pilot launch
References
https://hbswk.hbs.edu/item/minorities-who-whiten-job-resumes-get-more-interviews https://www.forbes.com/sites/janicegassam/2020/02/20/are-job-candidates-still-being-penalized-for-having-ghetto-names/?sh=432d831250ed https://www.marketplace.org/2021/08/03/new-research-shows-racial-discrimination-in-hiring-is-still-happening-at-the-earliest-stages/ https://www.livecareer.com/resources/careers/planning/black-job-seekers-face-racial-bias-in-hiring-process
Domain.com
equalizewith.tech
Built With
- react
- tailwindcss
Log in or sign up for Devpost to join the conversation.