An Intro to Unbiasify
Written in October 2020
A few years ago, I had a recruiter reach out to me on Twitter to discuss a project. Unlike the usual “Plz come work for my company” pitch that one would expect - instead, he was responding to a post I’d made about seeking open source projects for the TorontoJS meetup group to contribute to during a hack night. This was what I received:
I was intrigued
One of my goals in the development world is to
Make the world a better place through software - this sounded like a perfect fit with that goal. Long story short, Martin and I met for a coffee and ended up discussing this tool that had begun inception with some former colleagues and other contacts in Toronto with the help of StackAdapt (Martin’s employer at the time). After a quick chat, I was sold and knew that not only was this a project I cared about, but it was a project I wanted to help succeed if I could.
Unbiasify in a nutshell
The Unbiasify site sums it up pretty well:
Unbiasify hides names and profile photos on social networks so you can focus on what actually matters when recruiting and sourcing candidates.
The way I normally describe it to those who are interested is:
Unbiasify is a tool that aims to reduce unconscious bias in recruiters. We do this by obfuscating names and photos of applicants so that recruiters can focus on an applicant's qualifications instead of the information that has nothing to do with their abilities.
Tech and Stem in North America suffer from a lack of equal representation in our workforces.. That’s not an opinion, it’s a fact. It’s a fact that is apparent when you visit most tech related offices. It’s a fact that needs to change. I hope that through tools like Unbiasify and other training, we can help to influence change.
A swell in activity
After I joined the team, we began to support all sorts of networks with Unbiasify, received all sorts of contributions from the open source community (not just in Toronto!), and formed a team of core contributors.
Martin and other core contributer Bhavya Shah gave presentations explaining unconscious bias, why it’s a problem, and how Unbiasify could help. We were even mentioned in a couple of prominent online publications (you can find those on the Unbiasify site). We were excited with the progress and adoption, but our adoption was still very low.
After a storm of updates that included support for some of the biggest ASTs, due to no particular reason, the project slowly became inactive. The team’s lives got busier, employers changed, and Unbiasify unofficially went on the back burner.
Fast forward to 2020
2020 is/was a difficult year for so many reasons that I don’t think I need to go into them - everyone knows what’s happened in 2020. Mid-way through the year, with the re-emergence of the Black Lives Matter movement in the public spotlight, Justice, Equity, and Equality began to be something that many companies aimed to improve in their work places.
As recruiters are and have been aiming to improve their practices, we’ve seen Unbiasify usage increase at a rate we’ve never seen before. We’re not making headlines or anything like that - but there is a lot of growth in users and usage. And that’s encouraging.
With this in mind, the project has become busy again:
- Much of the original core group of contributors have returned to the project and are contributing in various ways
- We have already attained a partnership with a major player in the recruiting software space so that we can better support their platform, users and use cases
- We’re working to attain similar partnerships with the other players in the space
- And we’re improving the Unbiasify experience both for our users, but also for our developers by improving our contribution pipeline and updating our codebase.
I’ve started this post to explain a little bit what Unbiasify is and what it does, also to provide some history as to how we got to where we’re at. In my next few posts, I’m going to talk a bit about how it works as an extension and as a piece of code, what our development experience and pipeline is like, and what’s next for Unbiasify. I’ll update this post with links to those various articles when they’re written.
Thanks for reading, be excellent to eachother 👋