To celebrate International Women’s Day and Women’s History Month in March, I’m interviewing 4 exceptional female professionals who I admire to talk about topics on career development, leadership and gender equality in STEM.
My first guest is Dr. Yan, AI and Machine Learning Architect with over 15 years’ experience in the biotech industry. Yan shared her experience starting in this interdisciplinary field more than two decades ago and growing her expertise ever since. We also touched on topics like work / life balance as a working parent before and during the Pandemic, how to improve the ratio of women in management and more.
You can find a summary of our conversations in this blog post.
For the complete interview, please watch this video on Yan’s YouTube channel here.
Sinan: Please tell us who you are and what you do professionally.
Yan: I’m a Machine Learning and A.I. Architect at a biotech company in the U.S. I’m currently leading this project to build a system utilizing machine learning and deep learning approaches to make predictions for recommending certain prescriptions, e.g. antibiotics, with the right amount of dose for patients.
In the past, I worked on medical imaging and image analysis for more than a decade.
Sinan: How did you get started in the field of biotech / medical imaging?
Yan: It has evolved for sure, but it hasn’t changed drastically either. I’ve been following this path ever since I was in high school. I felt like I was going to be an engineer. That’s just in my mind all the time.
Sinan: Where did that come from?
Yan: I’ve never been the very best student in my class. The top students were going into science, like pure science, mathematics and physics. But for me, I’m good enough at learning new things and applying new knowledge, but maybe not at a very theoretical level. I’m more like an engineer type. So that got me into engineering.
Even in engineering, my grades were not the best of the best. At that time, if you went into engineering, you would go into computer science, electrical engineering and electronics. Those were the majors the top-tier students would go to. As for me, I wanted to find my niche – like a sweet spot that I would be good enough to stay in to find my path. At the same time, it might not be super challenging.
That’s why I started my professional career, or even before that – choosing a college major – going towards this interdisciplinary field, biomedical engineering. It combines medicine and applied engineering approaches to solve a specific set of problems.
As it evolves, it seems it’s playing out very nicely for me. I see a lot of opportunities in this field, especially in the past 10 to 20 years with the new development and tools from the engineering field. You can apply a lot of them in the medical field and biotech.
Sinan: It’s really early for you to have realized that you need to find a niche to be competitive and to have more opportunities in your career. I think it’s really important for young people to think in that direction too, instead of just competing with a single skill set.
Yan: In the career playground, you are going to discover a lot of these types of positions. They’re professionals, but they’re not really sitting in any of those big lists of majors for college applications. You don’t see those roles or skills in the major listing. But they are real jobs and they have real values. So don’t limit yourself in a small box where you want to fit into one single descriptor.
You can stack your skills as well. You can have knowledge in medicine, for instance, and knowledge in computer science. Stack them and then you’ll be able to find something in particular in there.
Sinan: How did you accumulate your knowledge in medicine?
Yan: It’s also evolved over time. Even though I went to school as a biomedical engineering student, we only had very basic courses in medicine, like basic anatomy and pathology.
You have to use almost the other part of your mind to learn that piece of knowledge, because we still have very limited understanding of how biology works. We’ve created computers. We design computers, we know how they work, but we didn’t design ourselves. Our human body is still a mystery to ourselves. I gave birth to a child, but I didn’t create them at all. I don’t know how they were created. It’s such a mystery and it’s so fascinating at the same time. So I think we should go into medicine and biology with such a modest mindset, the discovery mindset instead of the learning mindset.
During this process, I would like to try to discover for myself what my observations are and how I can apply my knowledge and observations to those problems I’m solving right now. I haven’t really gone through formal medicine training or gone to medical school. I’ve accumulated lots of knowledge while working on specific projects. Recently, I’ve learned a lot about bacteria and antibiotics, how they interact with each other, how they work at a microscopic level. They are really fascinating to me.
Sinan: I’d like to talk a bit more about career development. You have experience for over a decade. What would you say was the biggest challenge in your career?
Yan: I think it’s like a whole life challenge. In the past 10 years in my late twenties to thirties, trying to combine everything, every aspect in life, like having children, starting a family and at the same time carrying on your career – for me work/life balance was the most challenging piece.
You have to accept what you’re facing right now. You have to accept the priorities that you set for yourself. Do you want to put more of your time and energy into pushing your career forward, or do you want to split out a little piece of you – not just a little piece, a big chunk of you – into having kids, nurturing them, creating an environment, and spending time with them. If you want to handle all of these things together, I think you have to accept that during a few years, your career might have to slow down a little bit.
I was lucky enough to not have quitted working, but I have to admit there were some very challenging times for me. Sometimes I wasn’t able to sleep at night and the next day I had to go to work, think about the project and operate as if I didn’t have kids. There was a period of time, career-wise I just stayed steady, or a little stuck. I wasn’t able to push forward. I wasn’t able to explore new opportunities. I was not able to answer calls from people who reached out to me with “I have this new opportunity. Would you like to connect?” I just said no. I had to say no to keep my life going.
That was the challenge I faced for about 5 years. It’s long, but I wouldn’t say it’s such a bad thing, though. During that period, I felt very grounded in my work. I told myself, I don’t have the capacity to expand right now, so why not just focus entirely on what’s in my hands and try to solve the problem, try to lay the ground, try to build the product very nicely.
Unexpectedly, building my foundation during that time gave me an edge later on. Even though you felt like you had to slow down, at the same time, you allowed yourself to take time to understand the whole system, to actually do the work, to build things without any rush. You’d be very grounded and solid.
Sinan: Did you make the decision to grow in the technical track consciously? Are you open to becoming a people manager in the future?
Yan: It came to me naturally. I’ve always been strong technical-wise, and I liked to be involved from a technical perspective. I wanted to spend more of my time solving the real problem instead of solving the relationship with people. That piece is very critical. We need someone in the team to work with individuals, to help them realize their best potential, and to work collaboratively. I absolutely acknowledge the importance of the management role. It’s just for myself, my curiosity lies more in the projects themselves.
Of course things can evolve to a point where I feel I could contribute more as a manager, or where I were to start something myself. I’ve always had this idea of potentially starting something like my own project or company in the future.
Sinan: We are all aware there is a much lower ratio of female leaders than male leaders, not just in STEM. What’s your experience, and what are your thoughts on it?
Yan: I happen to have worked for a company with a female CEO, and that was a really nice experience for me. Leadership-wise, she’s totally capable and she was also close to us. She was willing to come to me and talk to me individually. I appreciate that aspect of her leadership.
Going back to the entire situation of why we have less female leaders, I think it’s twofold.
One is the willingness from our perspective. If I ask myself, will I be able or am I willing to step up and be a leader. I would hesitate. I would question, do I have what it takes to be a leader? What if I embarrass myself in front of 200 people? I feel I need to pump myself up a lot to be in that position. However, watching the male leaders in the past, I see that they make mistakes. Sometimes they go up there and say things that are just stupid and people would even laugh at them. But that’s okay. They’re still in that position and they don’t get fired. Maybe they get fired a few years later. Then they go to a different company and they can still be the CEO. So that’s something really interesting for me to see.
Another aspect is the opportunities. If a board is looking for a CEO, will they be willing to hire a female? Of course, maybe when you look at your choices, all you have are five males. But what if you have a female in the picture, would you consider her? What are the things you’re going to consider? Are there positions that take females in without thinking they’re women, and just look at their capabilities?
So both things have to come together.
However, going back to the root of things, what tells the difference between a man and a woman is the ability to bear and give birth to children. That evolves eventually to all of these differences in the workforce. Is that a problem to solve? Maybe the best thing to do is to provide more solutions for everyone. I think eventually we want options and more ways to choose how we want to live our lives.
Sinan: Machine Learning is a field that’s evolving very fast, especially in the past 5-10 years. How do you stay up-to-date in the field? Have you ever felt anxious about having to constantly learn new things to stay relevant in ML?
Yan: Going back to the PhD training, you stay up-to-date by reading recent papers. Nowadays people are so much more willing to open source their work, including the paper, code, datasets and other tools. It’s a lot easier now to learn and contribute in this field, compared to when I was doing my PhD. I think everyone should try today’s free online educational resources first, instead of going back to school.
I wouldn’t say I’m anxious about having to constantly learn new skills or knowledge. Everyone has a history. How you used to approach and solve a problem is experience and skills that are not going to be a waste. Nowadays in deep learning, we talk about transfer learning. You train a model on a generic data set, then you train it with specific knowledge and labels. The initial learning effect is not lost. I think as human beings, we have that transfer learning ability as well. Everything you’ve built in the past, they’re still relevant, even though people have stopped using some of those tools. You have to pick up new tools, but all of that is to make your life easier.
Sinan: When it comes to career change, what kind of opportunities are you looking for, and how do you evaluate them?
Yan: I’ve held three positions at three different companies so far. For my first switch, it was because I felt like there’s not a very clear path for me to move forward. I wasn’t sure whether I was going towards the management track or going towards a higher level on the technical track. The team had stayed stable for quite some time. At the time, a previous coworker approached me saying, we have new funding at this new company, would you like to give it a try. That came pretty naturally for me. It’s very relevant and it’s a project that I was very interested in and wanted to solve. I worked on that project for a year, and I felt like I basically solved it.
During that time, a headhunter approached me and presented me with this new opportunity. It’s outside of what I’d been doing in the past many years. I wanted to see what’s out there, instead of limiting myself to only medical imaging. I wanted to see what deep learning could solve outside of imaging.
It turns out I was able to transfer a lot of my knowledge into this new role, and to solve new problems, which gave me confidence. I don’t have to limit myself to a specific domain, even though I found for myself what I’ve been thinking was my niche. Now I’m thinking that I don’t even have to limit myself to machine learning or deep learning. I could stand up even further to solve problems from a system level. Machine learning is still just a tool. It doesn’t have to be my entire field. My field could be defining problems from a higher level, e.g. from the product level, from the users’ level, from the marketing level. I could then work backwards to see what options I have in terms of tools to solve a problem for that use case.