Meet a Data Scientist: Lucy Zou

WiDS Puget Sound and Data Circles is excited to present the next entry in our series, “Meet a Data Scientist!”

“Meet a Data Scientist” is dedicated to recognizing the amazing women powering the Puget Sound area’s data science community, spotlighting their journey into the field, their incredible accomplishments, and the weighty challenges that they faced along the way. This lies at the heart of WiDS Puget Sound mission of inspiring women to enter the data science field by showcasing its many incredible role models.

Do you know any marvelous women in data science? Send us a tip here!

“As a Data Scientist, you can build a model that improves offline metrics by 20% but you’ll still have to convince your team that it’s effective and needs to be put on the website,” says Lucy Zou, a Machine Learning Scientist III at Expedia Group (EG) in Seattle, Washington, “And you have to adapt your material to different audiences.” 

Zou's words encapsulate the multifaceted nature of a data scientist's role in the field’s dynamic landscape.  She has worked in data science for three years and has seen her own models deployed live on EG’s website. These models are  an important shift in how the travel booking company engages with customers.  Let's delve deeper into Lucy's journey and glean insights from her education and career.

Like many in the field of data science, Zou did not start her career in machine learning. She’d grown up in Shanghai, China and had come to the US as a high school cultural exchange student, initially spending a year in Virginia and then a year in Palm Desert, California. In high school, she discovered working with numbers was especially rewarding. After graduating, she left the 120 degree temperatures and sudden sandstorms of Palm Desert to return east, where she earned degrees in Accounting and Finance, and minored in Mathematics, at Georgetown University in Washington DC. 

Shortly after graduation, Zou began her career at Deloitte, where she worked on a Risk Advisory Team in Washington DC, consulting clients on IT database applications. This experience helped cultivate Zou’s communication skills, which allow her to engage successfully with diverse stakeholders today. After a year at Deloitte, Zou returned to university to pursue a master’s degree in Computational Analytics at Georgia Institute of Technology in Atlanta. She sought to merge her passions for technology and business with mathematics, and was excited to dip her toes into Deep Learning and Natural Language Processing.

During her studies, Zou gained valuable experience through internships in Strategic Analytics at Intercontinental Exchange. Shortly after graduating with her master’s degree, she accepted a position at EG and moved west to Seattle. She is glad she pivoted toward data science and she says it shows that it’s never too late to change what you’re doing if something else really sparks your interest. 

At EG, Zou is part of the Recommendations Team, which is responsible for implementing machine learning models to provide personalized recommendations to customers. She is instrumental in improving EG's recommendation system. The company is moving away from older models that relied on popularity to recommend cars and hotels, to more complex models that factor in user click-history, geographic area of interest, traveler preferences and other variables. These new models provide a more seamless and personalized user experience. They use embeddings, matrix factorization, tree-based algorithms and neural networks to now recommend new properties, destinations, flight bundles and activities that touch on different parts of a traveler’s journey.

Like many in the data science field, Zou works closely with a variety of stakeholders including product managers, platform engineers, machine learning engineers, and business stakeholders. “It’s really collaborative,” she says. “I love it and I’m really passionate about it. The collaboration is one of the main reasons I wanted to go into this field.” 

One of Zou's key contributions to her team has been her ability to visualize and communicate the impact of their machine learning models effectively. By using maps to illustrate how recommendations adapt to user preferences in real-time, she has been able to demonstrate model uplift and garner support for their implementation.

Zou is committed to continuous learning and development, which is also part of EG’s company culture. She takes advantage of opportunities such as weekly paper reading sessions with colleagues, as well as the monthly Day of Learning where the Data &AI team’s employees are encouraged to set down their work and invest time in learning new concepts and technologies. “It’s possible to read through a paper in 20 minutes, but it takes longer to really understand it. The Day of Learning is good for taking time to work with new concepts.” 

She also stays informed through research papers, blogs (she’s a fan of The Batch by Deep Learning AI), and online courses, like Udemy and Coursera. Many companies, including EG, offer access to learning platforms such as these, and Zou takes advantage of these benefits.  She most recently earned a certificate in Deep Learning through Coursera. Zou says that YouTube tutorials are another great resource for digging into specific topics.

Zou’s high degree of proficiency in communicating complex ideas does not limit her from refining her communication skills, which are important in data science. She emphasizes that practice is key, and she says that continuing to improve presentation and communication abilities is an ongoing process. Every bit helps. To this end, she actively participates in conferences, where she also gets a chance to learn about new models, best practices, and trends in academia and industry. Notably, she attended Recsys, a conference dedicated to recommender systems, where she gained insights into how other companies integrate recommender systems in their day to day operations.  

Zou places a high value on mentoring and knowledge-sharing as integral components of career development. By mentoring interns at EG through its Buddy program and actively seeking opportunities to learn from her colleagues, she fosters a culture of collaboration and continuous learning. Recognizing the importance of developing management skills early in her career, Zou emphasizes the role of mentorship in her professional growth, underscoring its significance in shaping her journey as a data scientist. These opportunities for greater responsibility help Zou to stay more organized, manage timelines, set clear tangible deliverables, and increase communicativeness about updates to her team. She’s very much an advocate for frequent and timely communications about updates to models and project details. 

Looking ahead, Zou sees team management as a potential path in her career but remains open to exploring different opportunities. She recognizes that, as an individual contributor, she loves finding solutions for different use-cases, reading research papers, exploring cutting-edge machine learning techniques, and seeing what colleagues are doing. She is aware that many data scientists come to a point where they make the decision to focus mostly on being individual contributors or becoming people managers. “It’s hard to make that decision if you don’t know enough about both roles, so having opportunities to learn about both will help you decide well when that time comes.” And if others are interested in management experiences, Zou encourages them to get a good taste of what it is like to be a people manager. To gain these experiences, Zou encourages others to talk to their higher-ups, voice the desire to take on more ownership of projects, more responsibility, and additional mentoring opportunities. 

When asked how she describes data science, Zou says, “Data is all around us. If someone asked me 5-10 years ago what data science was, I would have answered it is a way to scientifically analyze data. Now I know it is so much more. For instance, when you go into a store, data informs where products are placed, how they are selected, how many are stocked and how they are priced.” She says that generative AI (like ChatGPT) has been a disruptive force recently. It has been adopted widely and its development is happening very rapidly, which has helped many more companies personalize customer experiences, including small businesses. “There is a lot of potential for boosted business revenue, but that’s not the extent of it.” She says, “Data science is a way to extract meaningful insight and knowledge from many different scenarios, from structured and unstructured data, identify trends, behaviors, and risk factors and help everyone make data-driven decisions in competitive landscapes. It really is a way for us to understand the world around us.”

As an enthusiastic traveler, Zou is happy to work for EG because it offers her opportunities to explore new destinations and experiences. For her next non-data science adventure, she’s hoping to secure tickets and travel to a EuroCup soccer game in Germany this summer.