Women of Data: the data science team at Commonwealth Care Alliance®

Jill Hardy, Blog Manager, Looker at Google Cloud

Mar 22, 2021

In observance of Women’s History Month, I’m thrilled to introduce our latest Women of Data. This post is a special one because it features not only one, but three women in data careers with backgrounds in data science, statistics, and computer science — and they all just so happen to work together, making up the majority of the data science team at Commonwealth Care Alliance (CCA).

I’m excited to share the insights of Kaitlin Henry, Technical Success Manager; Marilyn McCallman, Technical Success Manager; and Shravanthi Sridhar, Data Scientist.

Becoming data scientists

Hello Kaitlin, Marilyn, and Shravanthi! Can you each tell us a bit about your background, and how it led you to a career in data?

Shravanthi: I was in my third year doing my bachelor's in computer science when I started hearing a lot about data & AI. I was intrigued to know more. I started slowly by taking small courses, which eventually led to getting my master’s in machine learning. I ended up working with a variety of data, but always had a soft spot for healthcare, where I dreamed of giving back to society by helping people.

My journey with CCA started as a co-op/internship back in summer 2019, which was then converted to full-time. Working with CCA has been exciting. I get to develop various models to help clinicians and staff discover valuable information from data. It feels amazing and boosts you up when you get to use your knowledge to help people.

Marilyn: I started out in the film and television industry. Eventually, I got burnt out on trying to compete in such a toxic environment, especially for women. That experience left me wanting to find a way to help others avoid ending up in similar situations. I pursued a degree in higher education and started counseling students on how to achieve their academic and career aspirations. From there it was a small jump to healthcare, supporting staff and patients alike in achieving their goals. I didn’t realize this was what I was doing at the time, but I created several libraries and databases to make information easier to access. So many people who need help, especially around social services, miss out simply because they don’t know how to or don’t have the means to access information to get help. Eventually this work led to the position I’m in now where I use my experience from working with patients and clinicians to design tools to enhance the work we do.

Kaitlin: I graduated with a BA from the University of Michigan and then pursued my master’s in Public Health, concentrating in epidemiology. I knew that health care was my interest area and thought that research and data could suit my skills. I meandered a bit career-wise afterward, trying to find the role that was the illusive "right fit," but trying to use data effectively was the undercurrent in all of my jobs after grad school. I was working as a Program Coordinator at CCA when I found an amazing opportunity on the data science team. The team was producing incredible tools but needed someone to help make it all palatable and usable for the clinical staff, so I became the bridge and felt like I finally hit my stride.

What it’s like working in data

What is one of the most impactful ways you see data affecting the workplace today?

Shravanthi: I see data-informed decision making without human bias as the most impactful way data is affecting the workplace.

Marilyn: Data fluency seems to be quickly gaining a foothold as a competency in organizations across industries. In healthcare, we’re seeing ways to use data to find better ways to address social and mental health needs. Thinking about how our social workers and health outreach workers can interpret dashboards and reports to identify people who need help before they even ask for it is pretty wild to think about.

Are there paths that you may not have taken if not for data to support your decision?

Shravanthi: In my role, decision making has always been supported by a combination of clinical context and data. Dealing with healthcare data, we always make sure to go beyond the data and analyze the results with context as well before making any decisions! In the same way, we also tend to prefer building models with higher explainability and try to avoid black box models.

Kaitlin: I'm not sure I would have attended a four year university at all if I hadn't spent a high school summer poring over college admissions data and statistics about graduates!

What has it been like to work on a data team comprised mostly of women?

Shravanthi: Ours is an amazing multidisciplinary team where we not only have majority women, but also each of them come from a very different background. Together with experience in clinical, epidemiology, statistics and data science, we get to learn a lot from one another. Personally, being a person having a strong CS background but not a strong clinical one, my team has helped me a lot in understanding the healthcare processes and data.

Marilyn: This team is certainly an anomaly and I’m so grateful every day to be a part of it. To work with such collaborative, brilliant ladies who want to help and support each other is a dream come true. I have learned so much from both Kaitlin and Shravanthi in such a short period of time. I wouldn’t be where I am without them, and our organization wouldn’t be benefitting from our innovations had we not been able to work together in such a seamless way. We’ve accomplished so much by working together.

Kaitlin: I’m so proud to be on a team with diverse expertise and a supportive, collaborative spirit. I have learned a lot from both Shravanthi and Marilyn and I think we’ve cultivated an environment that makes data approachable and even fun for the rest of the organization.

Advice to women who want to follow in their footsteps

Have there been any words of advice that you’ve kept top of mind as you’ve grown throughout your career?

Shravanthi: My advice to myself has always been to first understand the data before jumping to conclusions. Always spend that extra time to better understand the underlying patterns. Another thing that has always helped me is to think beyond just the objective when you look at the data. This has helped me in identifying connections between projects and improving models, and achieves better results.

Another piece of advice has always been to consider if the problem is worth solving, by analyzing the actions that will result from the output, which we refer to as the output action pair (OAP). In addition, this also gives us a clear understanding of the utility of models.

Marilyn: Big risk, big reward! I didn’t even know my current role existed until a couple years ago. Changing careers, trying something that’s never been done before, building something you have no prior experience with can be really scary. If I hadn’t done any of this, I wouldn’t be here right now. Allowing yourself to be uncomfortable and vulnerable can feel terrible, but it can also work wonders for you if things do work out.

Kaitlin: I hope "fake it til you make it" isn't too clichéd. It's been really freeing to start to push through the imposter syndrome and realize that most people are learning as they go and taking opportunities as they come (and that that's a good thing). I've picked up a lot of odd skills through independent learning because I didn’t want anyone to know I was faking it!

What advice would you give to other women who are interested in pursuing a similar career path to yours?

Shravanthi: Spend time in strengthening your foundation. Keep learning, as this is a field where new updates come up on a daily basis. Make sure you truly find and choose a career that you love.

Marilyn: You don’t need to be formally trained with a degree to work in this field, although it helps. A desire and commitment to learn, and domain knowledge can go far . You can do it. Find a mentor to help point you in the direction of the right things to learn. Don’t listen to anyone who says otherwise.

Kaitlin: I would encourage anyone to build off their existing interests on their journey into data. There is huge value in understanding the context and business use for a data need and being able to “speak both languages,” bridging between the data and the customer/business user.

If you could tell your younger self one thing based on the experiences that have led you here today, what would you say?

Shravanthi: Stay focused and keep learning! Take time to go deeper in your learning!

Marilyn: Take chances and own them. If you don’t, someone else will. (I’d probably wear a t-shirt that says “MARILYN: take more math classes” to maximize the productivity of this visit).

Kaitlin: Stop wallowing in self-doubt and do cool things. And wear sunscreen.