CV

Last updated: March 4, 2024

To download a PDF copy of my CV, click here.

Education

• B.S. in Applied Mathematics, University of California Los Angeles (UCLA), June 2024 (expected)
  Statistics Minor, Specialization in Computing
• Cumulative GPA: 4.00
• Relevant Coursework: Linear Algebra, Optimization, Machine Learning, Algorithms, Mathematical Modeling, Differential Equations, Numerical Analysis, Discrete Structures, Real Analysis, Complex Analysis

Work experience

• Data Science Intern, Snowflake (June-September 2023)
  • Developed adaptive stratified survey sampling framework with multivariate testing, scheduled via Airflow, to iteratively calibrate and refine user experience models with human feedback
  • Formulated mathematical framework for computing + updating account reputation scores via a combination of heuristics and behavioral anomaly detection
  • Preprocessed and analyzed third-party data classifications, investigated use of crowdsourcing label aggregation algorithms to improve internal data labeling system
• Pic 16B Reader: Python with Applications II, UCLA (January-March 2024)

Organizational experience

• Data Science Union (DSU)
  • Built and trained decoder transformer models in parallel from scratch on TinyStories, investigated scaling laws of dataset and model size with validation loss and created story generation demo
  • Designed a transformer attention model leveraging patent long-form text to classify new patents into USPC categories and produce technology forecasts, achieving a top-5 accuracy of 81.6%
• DataRes
  • Led a PageRank-centrality analysis graph project via Neo4j Graph Data Science and Cypher on 1,000,000 Spotify playlists
  • Augmented a message-passing graph convolutional network with custom-defined socioeconomic indicators to improve traffic accident predictions
• Association of Computing Machinery (ACM) AI
  • Developed a CNN in PyTorch to classify plant diseases
  • Built a bidirectional LSTM with GloVe embeddings to identify insincere questions on Quora

Projects and competitions

• 2022 DataFest Finalist + 2023
  • Cleaned proprietary data, derived and presented insights in teams of five from challenging long-form datasets (100+ columns, 2+ million rows) in 40 hours
  • Substantial use of seaborn and plotly visualizations, statistical tests, time series, survival analysis
• LLMs for Question Answering
  • Fine-tuned encoder-decoder model T5 with LoRA for extractive question answering on reading comprehension dataset SQuAD v1.1
• Art Generation with GANs
  • Implemented and compared DCGANs and Creative Adversarial Network (CAN)s to generate paintings, performed hyperparameter tuning and metric evaluations, developed interactive Streamlit demo
• Rocket League E-sports Statistical Analysis
  • Extracted 37,000+ series from public API, performed context-informed data wrangling and cleaning in pandas to obtain clear stat sheets for each player in every match
  • Generated exploratory visualizations with seaborn, identifying interesting correlations to investigate
  • Performed modeling/clustering to further analyze player behavior and uncover playstyle/team strategy insights
• UCLA Hack on the Hill 9 (2022) [Education Category Winner]
  • Designed skeleton framework for a novel UCLA automatic degree planner with a team in 12 hours
  • Wrote web-scraping algorithms incorporating regular expression matching to extract nested major requirements and prerequisite class data from various department and course catalogs

Skills

• Programming Languages: Python (pandas, NumPy, PyTorch, TensorFlow, Huggingface, scikit-learn), R, SQL, C++, HTML/CSS
• Technical Skills: Git, Airflow, Snowflake, Streamlit, Tableau, Docker, NLP, A/B testing
• Languages: English (Native fluency), Chinese (Native fluency)