“Meet Grand Rounds” is a series of interviews introducing some of the people working to make quality healthcare accessible to everyone, everywhere. The Data Science team at Grand Rounds builds algorithms that match people with the most appropriate and skilled doctors for their needs. The team constantly works to better understand our members’ needs as well as physician skill and expertise using data describing billions of historical clinical interactions. Meet the head of our Data Science team, Jayodita Sanghvi, in the Q&A below.
You were a finalist for the Silicon Valley Data Leader of the Year by the Women in IT initiative. Congratulations! How does it feel to be recognized?
It’s an honor to be recognized as a leader in data science. Healthcare data is inherently complicated. There are many challenges in harnessing this data and making sense of it so that it’s actionable for members, employers and providers. I’m proud of the work our team is doing to make healthcare more intelligent and more personalized, for people’s healthcare needs big and small.
Tell us how you discovered Grand Rounds, what excited you about joining the company?
I found Grand Rounds a little randomly through a friend. Prior to Grand Rounds, I was an academic, building computational simulations of complex living systems. While my work certainly had an impact on human health, the road to seeing that impact was very, very long.
One of my first conversations with Grand Rounds was with our co-founder, Dr. Rusty Hofmann. Hearing an academic tell me that the path by which I’d be able to drive maximal impact from the skills that I had was in industry and particularly at a company like Grand Rounds, was very compelling to me. He was absolutely right — very early on I found that I loved the fast-paced environment, getting to wear a lot of hats, learning from different people of different backgrounds, and most of all seeing the day-to-day impact of my work. There is something very exciting and rewarding about hearing and seeing the impact on our members through their stories in real time. And as a result, I haven’t looked back on the decision since!
Recent data shows that job openings in data science have increased 650% from 2012 to 2017. How did you make the transition from academics? Further, how did you make the transition to management?
I graduated from MIT with a degree in biology and minors in bioengineering and management science. I went on to get my Masters and PhD from Stanford University in bioengineering. I wanted to do research that would help human health and was excited by my academic work and its prospects. As my PhD work, the world’s first simulation of all of the inner workings of a living cell, was successful, I saw the impact that it could have on the field and continued similar research in my postdoc at UC Berkeley. As I attempted to scale my model, I realized that there was a lot in the field of “big data” that I needed to learn. I found Grand Rounds in my quest to find ways to learn Data Science. I also think I got really lucky that I chose to learn Data Science using healthcare data—possibly the messiest, noisiest, sparcest, and most error-filled data out there—because it pushed me even further technically.
Teaching was one of my favorite parts of my academic life, and in the industry setting, this translated to coaching and mentoring. This is what sparked my interest to grow as a manager. I had focused quite a bit of my academic time in business and management classes, but putting that into practice is a different ballgame. I am so fortunate to have had great mentors at Grand Rounds who have grown me as a manager and the support of my team who lets me experiment and grow with them. One of my primary goals is to accelerate the careers of each of my team members and help them achieve their goals. I work hard to keep finding new ways to do this.
Big data, data science and artificial intelligence (AI) are all popular buzzwords these days, but what do they really mean? Are these terms all hype or are they really solving problems?
The hype is real. Every single company is touting their AI, data science and deep learning skills these days. The promise is real, too, because we’ve seen many real advances that have been made across many industries.
For me, what matters more than the buzzwords and hype is the problem that you are trying to solve and whether you are using the methods that are most appropriate for that problem and the data that you have. My team works with the philosophy of starting simple and only adding complexity as merited. When the data you are working with is messy you have to be careful about the biases that may arise in choosing an inappropriate method. Moreover, simpler methods are often more interpretable in terms of back-calculating what inputs and features were most important in predicting your result. This interpretability makes conversations with physicians and clinicians easier, and in our work it is crucial to partner with those with medical background to make sure our model set-up and results are consistent with medical knowledge. Lastly, what is the problem that you are trying to solve? For us, regardless of simple or complex, everything we do is aimed at getting patients to better clinical outcomes, and that for me is motivating.
What excites you the most about your data work these days? What excites you when you think about the future of data?
Our product aims to help patients and healthcare providers make better decisions through the use of data. Our data sets can match a patient based on their specific needs, condition and preferences to a medical professional with the right level of expertise and experience to meet those needs. Being able to use data that patients normally don’t have access to, to make quality-driven connections between the patient and provider is what excites me in my day-to-day work.
Today, when someone has a healthcare need, it is extremely confusing, convoluted, and complicated to navigate the system. People don’t have any ability to get information about who has experience in the area of their specific need, and who can get them better clinical outcomes. I imagine a world in which healthcare isn’t an uphill battle, where routing decisions are personalized to the patients very specific needs and optimized for patient outcomes, and where everyone feels informed and confident in the healthcare decisions they are making for themselves and their families. I think creative uses of data can get us to a seamless, efficient, and high quality healthcare system.
Women in tech is a popular topic of conversation these days, particularly in STEM roles. How do you contribute to building women up when it comes to these type of positions?
It has always been a personal mission of mine to spread confidence to as many women as possible. Be it with the Boy and Girls Club that I worked closely with in graduate school, or the Women in Tech (WiT) community at Grand Rounds, growing youth and empowering women is and will always be a focus of mine.
Self-esteem and confidence is something that I have struggled with since I was very young, and from my own experience I know that it is something that can be grown and improved upon —especially if given the right mentorship, support, and environment. With confidence and a meritocratic environment in which all individuals have equal reach to all opportunities, women are unstoppable. Grand Rounds has been this for me — I have grown fast, and as I have demonstrated capability, they have opened doors for me. And as this has helped me grow my confidence, I want to share what I have learned so far (and keep learning) with with my team, the company, and my network in general.
Any words of wisdom for young folks graduating and hoping to dive into a data science-oriented role?
As you’re making a transition, regardless of whether you are coming from academia or industry, remember that you’re a highly trained person. Ask yourself: what do I want to apply that knowledge towards? What can I use my powers for, to have maximal impact in society?
What are you passionate about outside of work?
As of the last 20 months, most of my time outside of work is spent with my son. I am fortunate to have a supportive family that helps me feel like I am not compromising at work or at home. As a Data Scientist, I love watching him learn as there are so many parallels to how an algorithm learns. Everyday, I see him absorbing new inputs and evolving his predictions on how the world works. I hope he grows up in a world where every individual is given equal opportunities at work, where STEM fields are considered to be cool, and where high quality healthcare is easy to access.