Skip to main content Skip to secondary navigation

Q&A with Alumni: Claire Yang

Main content start
Claire Yang

April 26, 2022

Meet Claire Yang (MS '17), MS&E alumna and Data Scientist at Google.

Prior to her current role, she worked on the Enterprise Services side of Oracle as a Data Scientist. Claire graduated with a master’s degree in Management Science and Engineering from Stanford, where her area of concentration was Energy and the Environment.

“The things that have enabled me to be successful in my career are having a statistics and math background, and also having data intuition—basically a sense of how to approach a problem. For example, if you have a high-level business problem, how do you transform or convert that into an analytical problem, and what type of model or methodology can you propose to solve that problem? That is the core skill of being a data scientist.”

Why did you choose MS&E?

Stanford was my dream school. I originally completed my bachelor's degree in computer science, and during that time I realized that I was more interested in working on data-related projects, instead of just implementing things in code. So I was looking for a program that would best fit my interests and also help me develop the skills as a future data scientist—basically something that would combine engineering skills and management skills. That’s the reason why I came to MS&E.

I did some research before coming to MS&E, because I knew there were many different tracks or areas of concentration and I wanted to narrow down my research interest. I worked on a project as an undergrad that was focused on energy and policy, or rather energy economics. I knew that MS&E has two very well-known professors, John Weyant and James Sweeney, who are focused on energy economics. So I approached them and told them I was interested in doing some research projects with them while I was a master’s student in the department.

Soon after that, I joined more data-related projects, such as taking a class with Professor Ramesh Johari. The class was called Small Data—it’s now called Fundamentals of Data Science—and the year that I joined M&SE was the first year it was offered. It was very interesting and intensified my desire to work in this field. After that course, I worked on a lot more data projects, which led me to a career in data science.

What does it mean to be an MS&E alum?

I am extremely proud to be an MS&E alum, and I always felt very connected to the department while in school and even now. Every time I see new grads from MS&E, I always want to help them learn a little bit from my own lessons in the program and from my career path. I like to share advice and insights that I would have liked to have received when I was a student in MS&E. And that’s something that I hope to do for students in the program now, through this interview and other opportunities working with the department.

The alumni network also provides a sense of belonging, so whenever I see that MS&E had some news or new research or I see something in my news feed, I’m just very proud to be a part of MS&E.

How did your time at Stanford impact what you do now?

It made a strong impact; I don't think I would be where I am now without my time at Stanford. I spent a lot of time on my courses, especially the ones related to data. I mentioned Fundamentals of Data Science with Professor Johari, and there were also other classes related to probability and statistics. They were all very helpful.

The research projects that I participated in also contributed to my success. I mentioned one with John Weyant that was focused on energy, and through that I was pointed to another professor doing some natural-gas-related projects. I also worked on a few projects with professors in the medical school, processing data for their CT protocol . All of those were very interesting projects that sparked my interest in the field and also assisted me in defining my skills as a data scientist.

Was there a particular class or professor that really stuck with you?

Yes, in terms of the data-related skills, Ramesh Johari’s Fundamentals of Data Science course. It’s a really great class. Also, probability courses and Professor Yinyu Ye, plus the linear optimization and custom optimization courses. I also remember taking probability and advanced probability and risk analysis taught by Professor Ron Howard (now Professor Emeritus). I think all the hard skills classes are very insightful.

Did you have a professor that you saw as a mentor?

Yes, definitely. Professor John Weyant served as a mentor to me during my time here at Stanford. I did two years of research with him that focused more on the energy side, although my studies did not lead to a career in energy. That time with him helped me on my journey, and it was extremely helpful in the process of becoming who I am today.

Can you tell me more about the work you're doing now?

I'm currently a data scientist at Google Cloud, and my job can be divided into three parts. The first part is building machine learning models. My main focus is to build recommendations into the Google Cloud system that predict a user’s behavior based on historical usage data and profile data. For example, if we observe that a customer always under-utilizes one of their virtual machines or resources, we will recommend that they purchase a cheaper one because it’s a waste, especially if they have a large virtual machine that’s not being fully utilized, aiming to build  trust and establish long-term relationship with the customers.

The second part is to help the product team support their feature launches. Basically, there are a lot of features launched every year, so we need to quantify their impact and minimize their risk. We need to understand the tangible impact of a feature before we launch it broadly to all of our users. So we need to ask ourselves questions like, “What are the potential savings to the customer? What are the potential revenue impacts if we launched to other users around the world?”

The last part is community work. One example is interviews; we are constantly doing interview work to help Google recruit more qualified data scientists. Another part of the community work is building things—statistical packages, libraries, code—that can be used by all the data scientists at Google.

We have a Stanford alumni group within Google and I am one of its moderators. We meet for lunch on the last Thursday of every month, where we get together as alumni and talk about recent news and our thoughts about Google and Stanford. It’s just interesting to have this group at Google.

Is there a similar group for Berkeley and other universities at Google? Do they create their own?

I think it's pretty flexible, as long as people share an interest in developing a group. Right now, Stanford has over 2,500 alums participating in the Stanford Alumni Group at Google. There were a large number of employees interested worldwide, but in order to accommodate the lunch we only included the Bay Area for now. We do have a multi-lunch coming up for Europe and Asia, but it is not hosted by us, rather by the group coordinators there. For the Bay Area, it’s me and two other individuals, and all of our events are virtual at the moment.

What are one or two things that have enabled you to be successful in your career?

Well, the first thing is possessing the core skills, basically the technical skills. No matter what company you are interviewing with, the reality is in order to pass the interview, you need to have a solid background in the requirements for data science. It’s ideal to have a background in statistics or math, and also to have data intuition—basically, a sense of how to approach a problem. For example, if you have a high-level business problem, how do you transform or convert that into an analytical problem, and what type of model or methodology can you propose to solve that problem? That is the core skill of being a data scientist. Also, taking relevant classes and really asking the hard questions in class really helps with building your foundation.

The second thing that has helped me be successful is communication. Especially in the leadership work that I’ve done, communication was very important. And no matter which role you’re in, for data science or other technical roles in a tech company today, the success of a big feature depends on teamwork. No one can achieve large-scale projects on their own, so you need to communicate very well with your team members and any partnering teams involved in a particular project or feature.

Speaking of teamwork, I think it’s extremely important to network, get to know more people and get them involved in your group’s work. Back in 2015 when I first joined the department, there were a lot of opportunities to volunteer, to help organize alumni reunions and other events like that. Those were really great opportunities for students to get involved in that type of community work.

What advice would you give to current MS&E students?

Career planning is very helpful. Sometimes you might not know what you want to do right away, but I think you get there by doing research and explorations. And by taking a variety of courses, you can gain a better sense of what you want to do or achieve in the future. You might seek advice from experienced alumni or just search online for how to enter the field. There is a wealth of information online regarding how to start a career in data science, for example. And based on that information, you can figure what type of lessons or courses you need to gain the necessary technical skills.

Also, it can be valuable to read books beyond what’s required or take two classes on a particular topic, because you never know when you will apply the concepts or skills in your own reality. Just be like a sponge, soak up more, and a little more, and never stop learning. And I would say participate in more activities and be open to all opportunities that Stanford offers. Try to learn from others in those spaces.

Can you offer some words of wisdom for MS&E alumni and students interested in pursuing a career similar to yours?

If you’re interested in becoming a data scientist, there are three different core skills that you need to possess. The first one I already covered—you have to have a solid background in statistics or math.

The second part is data intuition, or product sense. This part might be more challenging, especially for the new grads or for students, because there isn’t a class to train you for what it’s like to be in industry. Not that Stanford or any other institution is lacking those classes; it’s just not something you learn in the classroom, but rather something you learn from your experience being in the position. For example, imagine we just released the new version of Android—how do we measure the success of that release? That type of problem won't be something you examine in a class, but rather you will learn from your own experience.

I would encourage students to think more about questions like this in their real lives. For example, whenever you use a mobile app or a website, ask yourself, “Why did they design it like this? Why does the layout look like this, or why did they change the button from the original to a new one? How would I measure the success of the change and what kind of metrics can be used to measure this particular feature?” Essentially, how would the developer know if a user likes this feature or not?

There won’t be many times when a product team or a company comes to ask you if you liked or didn’t like an update. And even if they do ask you, for example by offering a survey on a website, most of the time you don’t respond and you just find it annoying, right? But a company or a product team does need to have a way to quantify their impact or know whether their users are happy. That's something you can ask yourself—how are they collecting my data and how are they leveraging it? That type of thinking is like low hanging fruit—you don't have to pay for summer classes; you can just train yourself by asking some product questions like these ones.

The third skill for data scientists is coding, but this coding is not the same as the skills required for a software engineer. The coding for data scientists is more like data manipulation, data collection, feature engineering and data interpretation or modeling.

It’s pretty easy to improve your coding skills, because we have a lot of great courses not only in MS&E, but also in ICME and CS. For example, I remember I took Python 101; it was a pretty easy class for only one credit, and you could just take it as a seminar. There are good opportunities at Stanford for you to get familiar with coding and the various languages while getting some training and also training by yourself.

You can also develop some product cases and create hypothetical questions for yourself, such as: If I have a table of users who have upgraded to the latest version of IOS versus those that did not upgrade, what kind of metrics can I collect on their behavior? Now suppose I have that data, how should I analyze something like this? That’s something that we really do in our work as data scientists.

Is there anything you would like to say before we end?

My intention with this interview was to share advice and lessons learned with current students in an effort to help them to find a job, especially for those interested in a data science role. As I said previously, you really need to have a solid understanding of statistics, and there are some good classes like I mentioned—Professor Ramesh Johari’s Fundamentals of Data Science course, and probability and decision analysis courses as well. Beyond that, if you're interested, you can take some classes from the Statistics department or the CS department.

For the second part, the product piece, I didn’t recommend any courses because I didn't take them. I would encourage students to think about their real life experience every time they use a tech product.

And for the coding part there are some interesting and useful classes at Stanford, so my advice is to just explore them as you need. That should be more than enough to prepare you for a career in data science.

I would also advise students to have a plan when you are actively applying for jobs, and to also leverage the alumni network. I remember when I was applying for a job, I was able to acquire some help from the alumni; they gave me advice and also referred me to companies like Google. We have a strong network at Stanford, so make sure to leverage that because alumni, including myself, are willing to help students. My manager is also from Stanford, so I feel very connected in that way. And her advisor is also a woman, so we feel more connected and we are breaking barriers in terms of gender.

Be open and be positive, and try to get help from the people that you meet. Do not hesitate to reach out. It’s great to be open to new opportunities to volunteer work and to get to know more people. Be open to every opportunity. Your first job after you graduate will likely be hard because you're a new grad and you lack industry experience—so aim high, and fight harder. Don’t step back if you didn't acquire a big company offer; just keep trying and even if it’s a smaller company or start up, it might be a gold star for your career.