Data Scientist, Analytics
New York, NY, 1/17-Present
  • Lead product analytics for the Discovery and Search team. Define key metrics, design A/B tests and analyze data around features such as algorithmic dashboard streams and content recommendations. Conduct analyses to help decide what to build next.
  • Create ETL infrastructure for KPI reporting. Led development of a new framework for ETL, bringing modularization and unit testing into the Analytics code base for the first time.
  • Build analytics tools in general, such as reusable Jupyter notebooks for A/B test analysis.
Lead Data Scientist
New York, NY, 9/15-12/16
  • Managed and executed all data products from prototype to deployment, such as recommendations for creating content, machine learning-based tools for categorizing freelancers and proprietary scores for measuring audience engagement. Developed API service and worked with developers to design Ruby methods.
  • Built a recommender system to match Contently's content marketing clients with potential contributors from its freelancer network. Drove all stages of research and development: algorithm experimentation, offline validation, prototyping, user testing and algorithm evolution based on click data. Applied web development skills to transform the recommender from proof-of-concept to robust internal tool for Contently talent scouts.
  • Led efforts to make data-driven product development decisions: Analyzed product data to gain a better understanding of users. Synthesized findings to product managers. Emphasized data quality and consistent KPI measurement through consultations before development.
  • Contributed to a data-first culture across Contently by presenting data science findings and tools to stakeholders across the company and by undertaking data-intensive collaborations with services and sales team.
  • Mentored and managed a junior data scientist.
Product and Tech Lead
Melbourne, Australia, 7/14 - 8/16
  • Acted as product manager, designer and full-stack developer on an interdisciplinary technical-humanities team building Sonic Japan, a web application for sharing and visualizing soundscapes field-recorded around urban Japan.
  • Designed, built and maintained the application with Node, Express, MongoDB and JavaScript.
  • Continued to transform the application's design and project priorities based on analytics data. Defined goals and KPIs, planned/aligned with anthropologist team members and implemented features and redesigns in line with goals.
  • Manage content publishing schedule and distribution efforts to improve reach and engagement.
Back-End Engineer
New York, NY, 8/15-9/15
  • Project for a consumer financial startup: Python client wrapper library and mock server for the Jumio photo ID verification service using Flask, py.test, Docker, Jenkins.
Data Scientist (Part-time)
Melbourne, Australia, 3/15-2/16
  • Assisted in exploratory data analysis stages of a modeling project for a large financial SaaS company. Provided economic expertise, researched social network based methodologies, initiated feature selection.
  • Built web scraping systems for startup clients with BeautifulSoup and MongoDB.
  • Created visualization tools using d3.js, both as consultancy collateral and for custom client dashboards.
Senior Economic Analyst
New York, NY, 5/10-7/13
  • Covered Southeast Asian economies for the Emerging Markets group. Performed forecasting and quantitative analysis to develop a view on macroeconomic trends and risks.
  • Wrote briefings on economic, financial and political developments, forward-looking country risk overviews as well as analytical research pieces for Federal Reserve System-wide audiences and senior policymakers.
  • Collaborated with Asia team to create monthly regional presentations for senior management.
  • Managed all data for the Emerging Markets macro team. Sourced, aggregated, updated and manipulated datasets.
  • Presented data in create reports, dashboards and charts for presentations, briefing books and flagship publications.
  • Spearheaded the group's move from Excel spreadsheets to relational databases. Designed database, planned maintenance process, created new dashboards/reports and developed front-end interface for analyst team.
Market Intelligence Analyst
New York, NY, 2/10-4/10
  • Covered cloud computing at a boutique market research firm. Produced daily white-labeled reports as well as quarterly earnings analysis for corporate clients.
  • Wrote research reports on the industries of prospective clients.
CCG Private Equity
Analyst Intern
Shanghai, China, 9/09-11/09
  • Conducted industry research on prospective target sectors and wrote reports in English and Chinese for analysts and director, focusing on global commodity, European and American markets.
  • Created databases of Chinese companies for investment screening.
Master of Information Technology, Computing
Melbourne, Australia, 2013-2015
  • Honors: Graduated first-class honors average (86.6/100). Dean's list honors 2014-2015. International postgraduate scholarship winner.
  • Final software project: Natural language processing and data visualization for consumer decision-making and search.
Bachelor of Arts, Economics
Berkeley, CA, 2006-2009
  • GPA 3.92, graduated magna cum laude in three years, Phi Beta Kappa
  • Co-President, Fashion and Student Trends design club and Member/Newsletter Editor, Delta Phi Epsilon international affairs fraternity
  • Data analysis: Python (numpy/scipy, scikit-learn, nltk, pandas, gensim), d3.js, Excel
  • Data stores: Hadoop/Hive, Spark, MySQL, PostgreSQL, MongoDB, ElasticSearch
  • Web development: Flask, Node.js, jQuery/JavaScript, HTML/CSS, Git, AWS, Semaphore, Jenkins
  • Web scraping: Scrapy, requests, BeautifulSoup
  • Web analytics: Mixpanel, Google Analytics
Contently TalentMatch Built a recommender system to match Contently clients to potential contributors from Contently's network of 70,000+ freelancers. Experimented with algorithms and parameters; develop evaluation frameworks for offline validation; design, develop and deploy prototypes; run user testing and feedback sessions; track online user behavior to evolve algorithm continuously.
YMCA With a pro bono DataKind team, built an information extraction system for the YMCA national organization, with the aim of aiding measurement, benchmarking and analysis of YMCA branch operations.
No Free Lunch Built a tool using natural language processing and data visualization for consumer decision-making and search. Using a corpus of consumer product reviews from, one natural language processing module performed automatic extraction of important product "features" or aspects. Another performed sentiment analysis, with multiple methods including machine learning tested. This mined data formed the basis for an interactive, visually driven web app built in Node.js and d3.js. The app aims to bootstrap the knowledge transfer process by aiding analytical reasoning and search.
Search Engine Constructed a search engine following the vector space model. Based on evaluation metrics and manual error analysis of test queries, experimented with alternative normalization strategies, a biword index to support phrasal queries and a simple language detection strategy, and improved precision@5 by 20 percentage points.
Club-Biz Led a team of three to design and develop a social Rails app. Project encompassed front-end design (UX interviews, wireframes, UI mockups), architectural design communicated through UML and full implementation in Rails. Final web app included following, commenting, messaging, advertising, ticket reservation and reporting features.
Handwritten Digit Recognition Implemented an artificial neural network to recognize handwritten digits, plus a genetic algorithm to optimize neural network architecture.
Geolocating Users Developed a system for predicting user locations based on social graph data. Used Scikit-learn to preprocess data and test a range of machine learning algorithms for prediction tasks.
Streaming Video Created a multi-threaded client-server program in Java capable of broadcasting a webcam stream to multiple concurrent clients. Collaboratively designed architecture and coded in a team of two.
File Sync Wrote a multithreaded program for automatic file synchronization between a client and remote server.