Work Experience

 

Machine Learning Scientist – Monetization / User Demand / Pricing Strategy

May 2022 - Present Expedia Inc., Fortune 500

  • Led and collaborated with MLOps engineers on region-aware ML model by updating the pipeline and developing the margin ML model, leading to an +1.19% gross booking value (GBV) uplift and annualized +$21M dollars.

  • Led deep learning model developments for systems that outcome high-level demand prediction, contributing to an annualized +$237.5M uplift in Gross Booking Value (Q4 2023).

  • Built hotel-level ML pipeline for marketing budget allocation, resulting in +2.3% annual profit uplift.

  • Developed optimization toolkit using mixed-integer programming to improve pricing decisions at scale.

  • Generated 2.1% gross booking value uplift quarterly and 5.5 billion dollars revenue yearly via the newly-developed machine learning gross profit model with TensorFlow Lattice and LightGBM tree-based model.

  • Developed and utilized optimization package for decision making at scale to optimize pricing performance.

  • Created an automated data / training pipeline in collaboration with the ML engineers team to provide real-time updates of model.

 

Technical Project Manager - Monetization and Artificial Intelligence

June 2024 - August 2024 Expedia Inc., Fortune 500

  • Provided monthly progress reports to ML Science and Product leadership, influencing roadmap prioritization.

  • Collaborated across ML engineering, pricing operations, and product teams to ensure scalable pipeline delivery.

  • Mentored junior scientists on ML project design and model implementation; served as technical reviewer on pricing-related ML initiatives.

  • Supported strategic roadmap planning and coordinated cross-team ML initiatives.

 

Data Scientist Intern

2021-09 - 2021-11 Applied Materials

  • Built machine learning and deep learning models to predict the semiconductor production process behaviors, including wafer temperatures, power outputs, etc., that help automate the production process.

  • Adapted temporal convolutional network (TCN) and unsupervised learning models to time series to forecast the process parameters, significantly increased the average prediction accuracy.

  • Developed machine learning models on both the real-time and simulation data with PyTorch and SciKit Learn libraries.

 


Ph.D. Graduate Student

2017 - 2022 the University of Michigan

  • Published 10 journal articles during 4 years of Ph.D. career

  • Applied mathematical modeling skills and algorithms to multiple industry to facilitate solar cell and semiconductor designs

  • Served as a software developer of quantum software Q-Chem, working on computational methods by and collaborating with scientists for publishing 4 papers in material science and chemical physics.

  • Developed 3 new features of Restricted Active Space - Spin Flip (RAS2-SF) method in Q-Chem program in 3 years, with experiences using python, C++ and Fortran
    • Charge transfer states

    • Relativistic effects: spin-orbit coupling

    • Solvent effect: polarizable continuum model (PCM)

  • Created 2 Python programs used in optimization and process automation.

 

Education

 

Ph.D. Scientific Computing and Quantum Chemistry

2017 - 2022 the University of Michigan

  • Joint degree in Scientific Computing and Quantum Chemistry

  • Applied mathematical and computational tools to solve complicated physics and chemistry problems

  • Invented new algorithms to treate large matrix calculations

  • Awards and Test Scores

    • Department Research Fellowship (One-term fellowship awarded for top 6%)

    • PPG Summer Research Fellowship ($5,000 awarded for top 2% of program)

    • GPA: 3.7 / 4.0

    Relevant Courses

      Matrix Methods for Signal Processing, Data Analysis and Machine Learning, Numerical Linear Algebra, Computer Programming For Scientists and Engineers, Quantum Chemistry, Statistical Mechanics, Dynamics, SQL

 

 

B.S. Applied Mathematics and Physical Chemistry

2013 - 2017 Emory University

  • Double majored in Applied Mathematics and Chemistry

  • Studied with Prof. Joel M. Bowman and Prof. Alessandro Veneziani on applying computational and mathematical tools to solve complicated chemical and cardiovascular systems

  • Implemented simple heart rate / chemical kinetics with mathematical modeling skills using Matlab, Python and Fortran languages

  • Awards and Test Scores

    • Dean's List: 3 semesters

    • GRE quantitative: 170 (97th percentile)

    Relevant Courses

      Numerical Analysis, Linear Optimization, Nonlinear Optimization, Mathematical Modeling, Differential Equation, Partial Differential Equation, Multivariable Calculus, Math Foundation, Introduction to Computer Science (I. Data Structure, II. Algorithm), System Programming, Inorganic / Organic / Physical / Analytical Chemistry, Biochemistry, General Physics, Introduction to Accounting, Introduction to Biology