Pak Shing Ho

About Pak Shing Ho

I’m a data scientist with 8+ years of experience designing and scaling data science…

Experience

1. Applied Scientist Mle Economist

   Linkedin

   Jan 2025 — Present · San Francisco, CA, US

   I develop end-to-end machine learning and causal inference systems that combine experimental and observational data to optimize the trade-offs between rapid short-term experimentation and long-term strategic outcomes across a global user base of over 1 billion.• Build scalable causal ML & interpretable models to evaluate high-value user actions on key business metrics for 1B+ users.• Optimize A/B testing experimentation using the mixed of experimental and observational data and surrogates together with long-term outcome causal machine learning models to reduce dependence on extended experiments.• Embed time series outputs into operational workflows to inform leadership decisions and resource planning.• Lead cross-functional collaboration across Data Science, Product, and Finance to translate forecasts into action.Keywords: Machine Learning, causal inference, experimentation, time series forecasting, long-term uplift, engagement, retention, KPI, Python, Flyte.

Education

• Barcelona School of Economics

  Master's degree, Economics and Finance

  2013 — 2014

• Universitat Pompeu Fabra

  Doctor of Philosophy - PhD (transferred to Boston University), Economics, Finance and Management

  2014 — 2015

• University of Melbourne

  Bachelor of Commerce - BCom, Economics, Mathematics & Statistics

  2009 — 2011

• Harvard University

  Visiting Ph.D student, Economics

  2017 — 2018

• Boston University

  Doctor of Philosophy - PhD, Economics

  2015 — 2021

• UNSW

  Commerce

  2008 — 2009

• The London School of Economics and Political Science (LSE)

  MSc (Research), Economics

  2012 — 2013