It’s been a strange year, to say the least. Despite the total chaos, it’s a critical and exhilarating time to work in Artificial Intelligence. Whether you want to identify election fraud, understand climate change, or improve health care, machine learning is innovating all industries. The global machine learning market is projected to grow from $7.3B in 2020 to $30.6B in 2024, attaining a CAGR of 43%. AI-based processors, integrated memory and networking systems are projected to contribute a large percentage of market growth.
While these machine learning models strive to mirror and predict real life as closely as possible, the people behind these models do not represent the real world. Despite this rapid forecasted growth, women still only make up a measly 12% of the ML workforce. Only a quarter of all STEM workers are female, and even fewer are in positions of leadership. Despite a slight increase in the amount of women in STEM in recent years, the STEM gender gap persists as men enter STEM jobs at a more rapid pace than women.
Maintaining a balanced and diverse workforce is important to create an equitable technological future. As technology permeates all aspects of life and society — and will only continue to do so at a greater capacity than ever before, it’s critical to have a balanced and diverse workforce to think through complex issues that affect society at large. An imbalanced workplace puts teams at risk of translating the bias into the models they work with, and gender is just one of the many areas to tackle.
There is currently a lot of chatter about increasing gender diversity in technology. While growing gender diversity in the workplace is great, the scramble to hire for, and outwardly promote, a female-based presence in the tech industry can often feel inauthentic, disingenuous, and misused.
Instead of recognizing and reversing the systemic injustices women face, companies aim for the “diversity hire”, a stop-gap solution for the long term issues surrounding gender diversity as a whole.
The loud cry to hire more women and other underrepresented groups without making real systemic changes to the workplace and society at large is inimical to its own goals: these performative hiring practices yield the impression that women and other underrepresented groups may be hired to fill a quota or boost a statistic for investors, rather than for the hard work, intelligence, and perseverance that got them where they are. And this performativity is not lost on the new hires, who may be left with the impression that they are but a “diversity hire” — an impression that fosters feelings of imposter syndrome and self doubt that in turn perpetuate the low retention rates that the industry seeks to overcome.
Women who attain a job in tech, overcome the red tape, and speak about the adversity they’ve faced, become role models for women interested in the industry. They set an example of perseverance, professional and emotional intelligence, and outreach towards community — critical ingredients to a successful future in the industry.
The first step towards a healthier technological future is creating an environment where women feel empowered by their work and deserving of the positions they hold.
Role models are an essential part of this work.
By creating a sense of solidarity amongst women in (and interested in) tech, female role models help create an open conversation around the barriers, joys, and relationships they’ve encountered that helped them get where they are today. This conversation invites more women to share in the journey and pave their own into STEM, helps shift gendered perceptions of the industry, and can even catalyze institutional change around hiring norms and a real commitment to all forms of diversity.
So, I want to share some female role models with you.
I’ve gathered together female thought leaders in the AI/ML space who have paved the way for young females entering the workforce — and we asked them what they would say to their younger selves. From controlling interview anxiety to overcoming imposter syndrome, here’s what they have to say:
Rediet Abebe, Incoming Assistant Professor at UC Berkeley:
Throughout my undergraduate and large portions of my graduate career, I strongly felt that my ideas and perspectives that make me different were a weakness — that I should try to be as “typical” and conventional a researcher as possible to “make it.” This played out in a number of different ways from limiting the domains I drew research problems from to what techniques I felt I should prioritize.
It was during my second year of my Ph.D. work that I made a deliberate decision to do this work in a way that more-deeply resonated with me. My research now focuses on using algorithmic, computational, and mechanism design techniques to improve access to opportunity for historically disadvantaged communities. I co-organize MD4SG, a multi-institutional, multi-disciplinary research initiative on this topic, and Black in AI, a non-profit organization tackling equity issues in AI. I work with various government and non-government organizations, such as the NIH and community-based initiatives, to tackle problems facing marginalized communities. I approach my work more holistically so that different aspects of it — the research, teaching, mentoring, and equity work — are intertwined and inform one another. I have benefited greatly from this conscious decision that I made to define my own goals and be an AI researcher in a way that made the most sense to me, even if it looked atypical or non-standard. I would encourage any junior or aspiring researchers to do the same!
Inmar Givoni, Director of Engineering at Uber:
“It’s ok to be weird. It’s ok to be different. It’s ok to care about things other people don’t seem to find relevant or interesting or important. Being innovative and creative often means you aren’t interested in what everyone else is.
Don’t ask yourself if you are good enough to do X, ask yourself do you WANT to do X, is it exciting and interesting. If you are excited by it, you’ll figure out the rest. Oh, and a lot more people think they are imposters than not. It’s rare to not suffer from that. So, you are likely not special enough to be that one person who has imposter syndrome AND is an imposter.
Anima Anandkumar, Director of ML NVIDIA:
“Continue to nurture your love for mathematical foundations and augment it with knowledge of a variety of application domains. Develop communication skills. Be proactive in networking. Do not be too eager to join an organization without doing due diligence about its culture and values.”
Hoda Eydgahi, Head of Data Science at Brex:
“Lead with authentic confidence: ‘I don’t know, but I will find out’”
Pattie Maes, Lead of MIT Media Lab Fluid Interfaces:
“The name “Artificial Intelligence” is misleading and may conjure up images of HAL or robots gone rogue. But Machine Learning is really just a tool —albeit a very versatile and powerful one— that once you master it, you can deploy with major impact in any application domain you may feel passionate about, health, education, design, the arts, and more. Machine Learning is very accessible right now, some apps almost make it into child’s play, offering an easy point of entry into a field that keeps developing rapidly, is boundless and full of opportunities. If you want to change this world we live in, fight climate change, care for nature, improve education, make a difference in healthcare, I cannot think of a more powerful skill to master at this point in time.”
Amy Gershkoff Bolles, Chief Data Officer at Bitly:
“In my 15+ years in tech, I have faced significant barriers as a woman, including experiencing challenges ranging from micro-aggressions to outright harassment from senior male leaders in more than one job. When I bravely spoke up, I faced retaliation. Women must speak up when these incidents occur, no matter the cost, because only in speaking up can we begin to create a more fair and more equitable workplace. But we also must not let these negative experiences define us, and we must remember that the vast majority of people in the workforce are terrific allies, who deeply value diversity, equity, and inclusion.”
Dr. Radhika Dirks, CEO of xlabs.ai:
“My advice to you since you want to pursue a career in AI is this: DON’T. Instead, find the thing — the mission — that makes advancing AI a must for you. AI is an amplifier. You must have something to amplify. Very few people do. There are 1000s of incredible coders, so you should focus on something that in retrospect only you could have done! So, can you lean into a mission that truly inspires you and one you can’t seem to let go of? Something that makes you go cold or invades your dreams — like how 4 out of 10 people you know will be affected by cancer, or, can we emulate nature, or, how 7.6 billion intelligent beings with supercomputers and super wealth are still losing a battle against a tiny virus? Then become a vortex – a force of nature hell-bent on solving that problem. What you will see then is that AI or the right new technology will no longer be an option, but the only way forward. And in turn, you are no longer just an A-player in that technology, you will end up at the top because you can’t simply apply somebody else’s algorithms or approach to solve these big problems. Otherwise it would be done already!
Don’t ever lose that dazzle that makes you truly you. Bring it with you into the tech world.“
Bahar Sateli, Manager in AI at PwC Canada:
“Prepare for your job to the best of your ability, and then jump right in. Truth is, in a fast-paced and broad domain like AI, it is virtually impossible to master all relevant skills and stay on top of the latest advancements at all times. Pick a subdomain of AI that you are really interested in and focus on grasping an in-depth knowledge of its foundations. You will refine and master your practical skills while on the job. Apply for that position today, you’re as ready as you can be.
Claire Longo, Machine Learning Manager at Twilio Inc.:
“ML is a new field, so don’t expect the path to be paved for you when you join a new ML team. As an ML professional, you’ll be able to identify solutions to the challenges you face. Be willing to work outside the confines of your job description and use skill sets you’re not 100% comfortable with yet to help bring solutions!
The people you work with can have just as much an impact on your career as the company you work for or the products you help create. Choose a team where you can surround yourself with people who believe in you and in your ability to learn.”
Dimple Abani, AI/ML Product at Paypal:
“Take risks and don’t be afraid to fail. The best thing to do is to go out and try it. There is a lot of good information available online. Start with small projects. For example, there are YouTube tutorials that can show you how to build an image classifier in 5 minutes. To go deeper, you can take online courses, pursue a degree, find an internship, or join a startup. Be curious about how you can use AI to solve problems.”
Wendy Foster, Director of Data Science at Shopify:
“The advice I would give to my younger self, with regard to pursuing a career in AI/ML would be to never decenter the “human” in the work we do. Algorithms are not innocent, they encode the complexity of our experiences and assumptions (both good and bad), and in turn can inform and transform the experiences and assumptions of the end users of our implementations. Root your solutions in praxis over theory, and lead here – don’t follow.”
Katie Bauer, Data Science Manager:
“If you build it, they will come” is absolutely the wrong mindset. Building great stuff is table stakes, but just because you’ve made it doesn’t mean people will know about it or understand it. Knowing how to pitch yourself and your projects will open so many doors for you because it ensures your accomplishments get the attention they rightfully deserve. This can feel like bragging, but it shouldn’t. It’s self-advocacy! No one is more invested in your success than yourself, so don’t leave your fate in someone else’s hands.”
Timnit Gebru, Technical Lead of Google’s Ethical AI Team:
“I would tell myself not to worry that I took such a meandering path in the field, and tell my younger self that that experience would be one of the things that would shape my research direction. I would also tell myself that racism and sexism come from literally all angles, that everything I feel is real, and that each time I persevere I should be proud of myself and not minimize what I just went through.”
Allie K Miller, AWS US Head of AI Business Development:
“Machine learning thrives on the power of predictability, but our careers shouldn’t have to. Looking back, I would tell myself to embrace the randomness. The things in my life that are the most enriching now would have been the most impossible to guess 20, 10, or even 2 years ago. Don’t be so rigidly focused. A big goal can be a fantastic motivator, but it can also create blinders. It can create an illusion that there is this perfectly pre-planned, straight, narrow path to your goal. But meandering is healthy and invaluable. Picture your career path as a 14-lane wide highway, not a tightrope. Imagine infinite ways to get there, not just one. Apply predictability to your algorithms, but not in the construction of your life. Your “model deviations” might be the most important things you do.”
Erica Lee, Director of Machine Learning and Software Engineering at Zillow:
“Following my passion and surrounding myself with positive people have empowered me to work in AI & Software. At Zillow, I get to work on meaningful things with amazing people. If you’re searching for your next career move, follow your heart and find a great community you love!”