Thinking About Diversity: Bachelor Edition

This post is part of a series where we’ll explore how technology can change our perspective by writing and designing novel products every week this year.


Every couple of months, a conversation about diversity would come up, and I’d cite this story that I read a long time ago. Every time I mention it, I would search for hours trying to find it, but to no avail. I don’t know where it went, or why I can’t find it, but it changed my perspective forever. My butchered oral retelling is something like this:

I grew up as an inner city kid in the east coast. I always thought that Facebook is a cool website made by a few people, max 10. Everyone discouraged me from doodling in class all day, and that success is becoming a doctor. At some point I was part of a school trip where I mingled with adults. There, by chance, I bumped into someone who worked in Silicon Valley. Much to my surprise, that person told me that Facebook in fact hired thousands of people. Furthermore, some of them get paid six-figures to draw all day. He then told me about UI/UX design and its value. That random five minute conversation I had with a stranger changed the path I walked forever. Now I’m a designer at a large tech firm.

For me, that story made me realize that what I considered to be normal is only mine to own. It is not a guaranteed at all, that one would know, that designing for the screen is a highly lucrative endeavour.

I was privileged to be born into a family that valued education and encouraged exploration. My parents gave up the country they knew to move to America so I could have a better education. I knew the touch of a computer ever since I could remember. One should not be surprised that I ended up in Silicon Valley. Still, I too, would attribute my career today to a chance encounter.

Who would have thought this kid would grow up to be a designer…

When I was 8, I played a popular game called Neopets. The game is about cultivating virtual pets, and it had a feature that allowed you to trade with other players. Being an innocent time, you were able to add custom CSS and HTML to decorate your trading page. One of my closest friend at the time, Billy, had a cousin who was several years older than us. He ran a popular website that helped players customize their page with code. As an immigrant transferring to a new school, I thought this was the way to fit in. I started to learn to code and design (with Microsoft Paint!) and eventually made my own Neopets guide website. I have never met that cousin of his, but ever since that day, I have been designing and coding. I made countless themes for MySpace, Tumblr, Wordpress, and more.

Put it another way: about a year ago, I was talking to a friend who described how she had changed diapers as soon as her dexterity allowed since she had a little brother to take care off. Though it may seem silly, that moment was the first time in my life that I thought of babysitting as a tangible thing. As the youngest person in my family, I never needed to take care of anyone. My brother was close in age so he never babysat me. My parents has never hired a babysitter to take care of us (unless you count Mr. Rogers). All I knew about babysitting was as a plot device for movies.

For me, designing and coding is as normal to me as breathing. For another, babysitting was. Neither position is universal.

I can’t help but to think about the opportunities people have lost, or the skills the world has missed due to the fact that the we only know our corner of the world.

It’s hard to increase diversity when the people we’re looking for don’t even know we exist. Likely, the people who are looking for us, are people who had circumstances just like us.

Sure, we may have taken some test in high school that attempted to give us ideas of what we can be when we grow up, but how many of us took that seriously? In a world as complex as ours, how many jobs can a quiz actually contain? And even if we found a job that we do like, what do you do with that information if your immediate world knows nothing about it?

Absolutely not a scientific depiction of this issue.

Say what you want about dating apps, but it does a great job of introducing you to people outside your network. It does this by introducing you to many more people than you would encounter in real life. It allows you to go through them conveniently by providing small discrete digestible chunks: pictures, a short bio, a list of tags about what they’re looking for. Finally, it culminates with a button that leads to a real conversation where presumptions are broken down and both parties truly learn about each other.

Maybe we can swipe a page from their book.


In MatchWork, candidates create profiles that summarizes their work aptitude. These profiles are shown to mentors, people in the industry who wants to help others discover what they've found to be amazing. Mentors can use these profiles to consider the likelihood of the candidate’s success in their field. When a mentor finds a candidate whose interests and work aptitude aligns with their own, they can start a conversation with them: introducing them to the new field.

Example radar graph of people with different aptitudes. People who succeed at a certain career will likely have similar radar charts.

At the heart of MatchWork is a radar chart that summarizes, however roughly, one’s work aptitude. It’s true that we are flattening the candidate into a constrained dimension, but in return we get a glanceable visualization. In doing so, mentors can view a larger amount of profile, increasing the likelihood of a good match.

Although we can use machine learning and artificial intelligence to help match mentors with candidates, humans are still the best pattern detector. Using these radar charts, the mentors can start having a better feel for the type of individuals who might be successful in their given field.

Having a glance-able proxy for a larger profile is helpful when going through a large number of them.

By having a short profile with tons of glanceable information, viewers are able to consume many of them, thereby increasing the likelihood of finding the right candidate.

To provide additional insight to the candidate’s work profile, we provide multiple avenues of expression: personality, passion, skillset, consumption.

Candidates are allowed to choose three words to describe themselves, as well as a short bio about who they are. This linguistic based summary humanizes the radar graphs, while simultaneously giving the mentor a better understanding of the candidate’s personality.

The list of passion and skills provides a deeper look at where the candidate is currently. The list of passion provides insight into culture fit, while the list of skills highlights their ability to accomplish the tasks required for a certain career.

More interestingly, the list of things the candidate has consumed recently gives insight as to what the candidate aspires to be. This is important not only because we are shaped by the things we are surrounded by, but it can affirm or deny the accuracy of the profile.

Similar to “looking for” lists and integrations with apps like Instagram and Spotify on dating apps, these small, yet numerous, data points creates a constellation that helps the mentors get to know a candidate a little bit more.

Once a mentor finds a candidate that is likely to succeed in their field, they can dive into a more complete profile. The goal of the complete profile is to expand the spaces that were flattened by the short profile.

Similar to a cover letter, a longer space to write about who they are and why they should be considered is provided. Like a college application essay, this space allows the mentor to judge the candidate qualitatively. For example: someone who is affected by socio-economic injustice arguably understand the problem at a deeper, or at least different, level than those studying it externally —even though that injustice may actually be detrimental to their quantitative profile (e.g. school grades).

Next, candidates are able to add media to their profile. This is useful not only to lend credence through proofs to the profile, but also to give a visual break for the mentor. Often this can be in the form of a picture. For example: previous presentations for those who like leading, or previous art work for creatives. But it can also be in the form of quotes, demonstrating the way the candidate’s mind works — or even auditory, in the case of performers. It provides a free form place where candidates can express themselves beyond numbers.

Once we have narrowed down candidates, we can focus on learning about them in detail — beyond the listable quality.

Theses small bursts of information come together to paint a clearer picture of the candidate. Most importantly, however, the mentors are then encouraged to connect. Conversation is an important avenue for both the mentors and candidates to break down or affirm presumptions. The mentors learn about another’s story, while the candidates learn about a new field.

In addition, having a good conversation will provide the candidate with a foot in the door. People who grew up surrounded by a certain profession is more likely to be in that profession since they know what it’s like. Unlike other career finding platforms, MatchWork is centered around having conversations. Through these conversations, mentors can guide candidates like an older sibling would. Making it more likely for the candidates to seriously consider the new field.

Going through many profiles is not the goal, having a good conversation that leads to more is.

Say hi and get a high five for supporting others.

Design Details

MatchWork neither uses the candidate’s real name nor profile photos in order to discourage bias. Studies have shown that we are prone to biases. Academic papers with the same content, but authored by “John” instead of “Joan” are constantly rated higher. Papers written from Asian countries to a European journal is 5x more likely to be rejected.

The name given to profiles are generated by the app in order to humanize (vs. AnonymousUser123) while still protecting the candidates’ real identity and reduce implicit bias. Similarly, emoji as a profile picture is a good way to express one’s personality without having to reveal real identity because they are sundry, with each having their own expressive meaning. For example these emojis all represent different personalities: 🦁 😇 ✊ 😎 🤪 , yet all of them can relate to a field.

Another detail in the profile is exposing the current trajectory of the candidate. Allowing the candidate to list what they are currently working on encourages future self development. At the same time, this gives the mentor a vision of where the candidate wants to be. A more pragmatic way to say “in 5 years…”

By default, MatchWork gives mentors a number of candidates per day to go through that it thinks would be good matches. However, mentors may realize certain patterns that MatchWork has not yet learned.

Lists in profiles are not only good indicators for the current candidate, but they are also markers for other potential candidates. To help facilitate this, each item in the lists are clickable tags that link to more candidates that used those tags.


We can improve diversity by making sure the top of the funnel is diverse to begin with. Counter-intuitively, dating apps can give a good framework for connecting mentors with candidates. First, the more candidates we go through, the likelier it is for a mentor to find a good candidate. Then, they can progressively get to know the candidate better through discrete digestible information in an increasingly detailed profile. Finally, the most important step, is for the mentor and the candidate to connect through conversations so that they truly know each other. This not only expands what were flattened in profiles, but can also help get the candidate’s foot in the door.

p.s. If you know where that story from the beginning is, message me!

Don’t forget to subscribe to get notified when the next installment comes out!

Photo by rawpixel on Unsplash