Resume

Relevant Industry Experience

Meta Platforms
Research Scientist (ML Engineer), New York NY
Instagram Well-being Fairness Team
▪ Coordinate with partner teams to understand large scale machine learning systems, and measure/identify potential fairness or equity issues.
▪ Build mitigations using either machine learning or heuristic methods to minimize said issues, improve creator performance, and assist in regulatory readiness.

Creator ML
▪ Led projects with partner Creator teams to improve or build machine learning models / recommendation systems that improved both creator and viewer engagement.

News Tab Growth
▪ Focused on improving existing audience selection and promotional ad models resulting in both increase traffic and capacity wins to the tab.
▪ Led model maintenance after team dissolution and ensured long-term viability through recurrence.

Software Engineering Intern, New York NY
News Tab Recommendations
▪ Built interpretable heuristic-based user segmentation on News Tab based on unsupervised learning
▪ Investigated existence of potential bias in consumption and delivery on the News Tab
▪ Set-up user segmentation so that it is automated and malleable to future changes

Pandora Media, Science Intern - Listener Science, Oakland, CA
▪ Created & tuned recommender channel based on implicit signals & precision-weighted factorization
▪ Built a prototype that allowed systematically qualitatively and qualitatively testing of model iterations
▪ Set-up code for the recommender radio channel to be deployment-ready for live AB testing

Poshmark, Data Science Intern, Redwood City, CA
▪ Employed causal inference & ML methods to determine the economic efficacy of the site’s social features
▪ Examined the formation and evolution of users’ relationships & predicted if they resulted in purchases
▪ Explored community structures & tie formation locations through temporal analysis of the site’s network
▪ Conducted qualitative analyses of users’ experiences and their own comments – in order to devise strategic recommendations that could both benefit the goals of the company and its users

FCB New York, Senior Social Media Analyst, New York, NY
▪ Tracked efficacy of the FDA’s “The Real Cost” anti-smoking campaign across four social media channels
▪ Determined common topics on social media around campaign PSAs, vaping, and smoking in general
▪ Identified central influencers in conversations about smoking using network analysis
▪ Coordinated with social media marketing partners to effectively guide the FDA's social media strategy

Converseon Inc., Social Media Analyst, New York, NY
▪ Explored topics on social media for Fortune 500 companies including Twitter, IBM, and Nissan
▪ Identified key influencers across various markets using network analysis of social media data
▪ Managed a team of freelancers and Mechanical Turk users for data coding, and ensured reliability of that data

YPulse, Marketing Intern, New York, NY
▪ Devised, programmed, and analyzed youth-targeted surveys for major entertainment corporations
▪ Led community management with survey participants

Aug ‘21 – Present

Sept ’22 – Present
(Current Team)

Feb ‘22 - Sept ‘22
(Re-org)

Aug’ 21 - Feb ‘22
(Re-org)

June ‘20 – Aug -’20

May ‘19 – Aug ‘19

May ‘18 – Aug ‘18

Aug ’15 – Jul ‘16

Jul ’13 – Aug ‘13

Jul ’12 – Aug ‘12

Research Experience

Opinion cascades and the unpredictability of partisan polarization
Michael Macy, Natalie Tong*, Sebastian Deri*, Alexander Ruch* (*equal authorship)
▪ Funded by NSF Award #1756822
▪ PI: Michael Macy, collaborating with Alexander Ruch and Sebastian Deri
▪ Using the “multiple words” experimental design of Music Lab (Salganik, Dodds, Watts, 2006), we found that arbitrary & unpredictable correlations can occur between opposing political identities & beliefs/hobbies/lifestyles

Published in Aug ‘19

Finished in Dec ‘20

Conduits to the Alt-Right – Lifestyle drivers of Alt-Right participation on Reddit
▪ In collaboration with Michael Macy and Lillian Lee
▪ Exploring self-selection vs. influence in populist community participation on Reddit (especially among alt-right members) – did future populists hold populist ideas when they joined Reddit, or were they radicalized over time?
▪ If users are radicalized over time, does the mechanism primarily involve participation in increasingly-radical segments of lifestyle communities (e.g. gaming), or in increasingly-radical political communities?

Ongoing

Sept ‘16 – Dec ‘17

The impact of network topology and the navigation of norms on popularity on Reddit
▪ Undergraduate Thesis for the Rutgers University Sociology Honors Program, advised by Paul McLean
▪ Explored the interaction between network structure, norms, and popularity on the site, Reddit
▪ Observed differences in behavior between “one-hit wonder” and “celebrity” users
▪ Awarded the Henry Rutgers Scholar Award for the project

Sept ’12 – May ’13

The ethnically ambiguous generation: perceptions of interracial Families in television commercials
▪ Undergraduate Research Assistant for Crystal Bedley and Catherine Lee
▪ Explored the presence of ethnically ambiguous roles in ads and their efficacy in targeting minorities
▪  Gathered and coded data on uncertainty in the perception of race and ethnicity in TV and online ads

Sept ’12 – May ’13

A theme analysis of issues reported by adult children of alcoholics in online support forums
▪ Undergraduate Research Assistant for Marie Haverfield and Jennifer Theiss
▪ Identified and coded types of support in an online community for adult children of alcoholics (ACoAs)
▪ Found that ACoAs often expressed support in terms of solidarity – often focusing on seven themes (e.g. problems with romantic relationships, low self-esteem, or lack of a connection with one's inner-child)

Sept ’11 – Dec ’11

Other Experience

Computational Social Science Reading Group (Co-founder), Ithaca, NY

Cornell University LGBTQ Resource Center Mentorship Program (Mentor), Ithaca, NY

Summer School Series on Methods for Computational Social Science, Sardinia, Italy

Sept ‘18 – Jun ‘20

Sept ‘18 – Dec‘18

Jul '17

Honors and Awards

Cornell University
▪ NSF Award #1756822 (PI: Michael Macy): Testing Unpredictability with Multiple Worlds
▪ NSF Graduate Research Fellowship Program Award (2017-Present)
▪ Information Science Fellowship (2016-2017)

Rutgers University
▪ Graduated Summa Cum Laude
▪ Henry Rutgers Scholar Award (2013)
▪ Sociology Honors Program (2012-2013)
▪ The School of Arts & Sciences Academic Excellence Award (2011-2013)

Skills

▪ Programming Languages – Python, R, Scala, Hack, PHP
▪ Machine Learning Frameworks- PyTorch, TensorFlow, scikit-learn
▪ Database Languages – Presto, Hive, PostgreSQL, MySQL, Google BigQuery, NoSQL, MongoDB,
▪ Python Libraries – SciPy, NumPy, pandas, NLTK, PySpark, GraphFrames, Igraph, NetworkX, matplotlib, seaborn, plotly
▪ Experimental Design
▪ Grant Writing Experience
▪ Amazon Mechanical Turk
▪ Network Visualization Suites – Gephi, Cytoscape, Pajek
▪ Survey Creation Suites – Qualtrics, Globalpark EFS, SelectSurvey.NET, SurveyMonkey
▪ Spoken Languages - English (native), Shanghainese (elementary), Japanese (elementary), Spanish (elementary)