Dimensional Gateway Traversal Initiated
Dimensional Coordinates: Alpha/Omega/Prime Coordinates Locked
Initiate Quantum Calibration Singularity Detected
Beginning Tesseract Unfolding
Hyperdimensional Matrices Aligned
Traversing
Dimensional Shift
Quantum Entanglement Stabilized
Cosmic Strings Vibrating in Harmony
Wormhole Aperture Expanding
Dimensional Gateway Stabilizing
Reality Parameters Reconfigured
Quantum Fluctuation Nominal
Initiating Spacetime Fold
Scanning Parallel Realities
Analyzing Dark Matter Density
Processing Gravitational Waves
Calibrating Temporal Displacement
Evaluating Dimensional Resonance
Stabilizing Quantum Foam
Traversal Sequence Complete Dimensional Gateway Open

Model. Test. Trade.

End-to-end quant research, from alpha discovery to execution and risk.

Brightness

0%

Font size

14px

↑ Top

Education

Columbia University M.A. in Mathematics of Finance New York, NY

September 2024 – December 2025

GPA: 3.94/4.0

Coursework: Machine Learning, Time Series Modeling, Stochastic Process, Programming for Computational Finance, Numerical Methods, Derivatives Trading, Option Pricing.

Teaching Assistant: MATHGR5430 Machine Learning for Finance.

University of Liverpool B.Sc. in Mathematics with Finance Suzhou, China / Liverpool, UK

September 2020 – May 2024

GPA: 3.83/4.0 · First Class Honors (WES report)

Coursework: Stochastic Theory, Linear Statistical Models, Mathematical Risk Theory, Applied Probability, Derivative Securities, Numerical Methods, Operational Research, Statistics and Probability, Financial Reporting.

Experience

Numeraxial LLC Quantitative Research Intern New York, NY

July 2025 – Present
  • Defined cross-asset research questions and assembled point-in-time data, standardizing Bloomberg listed-equity, FX, corporate-action, identifier, calendar, and macro inputs with QA outputs.
  • Tested statistical, macro, and market hypotheses across HMM and clustering regimes, reviewing transition stability, economic rationale, and regime-specific drawdowns before portfolio construction.
  • Translated forecast evidence into portfolio implications by comparing mean–variance, Black–Litterman, and HRP under constraints, with Sharpe, drawdown, turnover, and risk-contribution review.

Western Securities Co., Ltd. Quantitative Analyst Intern Shanghai, China

January 2024 – July 2024
  • Established reproducible crypto hypothesis tests by encoding shared signal-state logic in a leakage-aware vectorized backtester for deterministic Alpaca replay across parameter runs.
  • Identified signal behavior and failure conditions for EMA–ADX trend and Z-score reversion gated by ADX and RSI through parameter-sensitivity, regime, turnover, and drawdown analysis.
  • Reported one-year in-sample research benchmarks (not live and not net of costs): trend Sharpe 2.13 and reversion Sharpe 2.65 versus 1.65 for buy-and-hold, supporting subsequent cost-sensitivity analysis.
  • Set research acceptance criteria by stress-testing fee and slippage assumptions, limit offsets, repricing, freshness gates, one-day parametric VaR, and drawdown envelopes.

Projects

SPY Economic Signal Research and Walk-Forward Validation CQF

June 2025
  • Tested whether volatility clustering, momentum, reversal, and seasonality explain next-day SPY direction using 40+ calendar-aligned features with timestamp and leakage controls.
  • Compared logistic regression, random forest, and XGBoost using time-series cross-validation, nested tuning, probability calibration, SHAP, and perturbation tests; achieved held-out ROC AUC of 0.714.
  • Mapped calibrated probabilities to long/flat signals, using walk-forward precision, recall, turnover, drawdown, and cost sensitivity to review threshold stability and failure conditions.

GitHub

Factor-Neutral Equity Portfolio Research and Attribution CU

December 2024
  • Built a factor-neutral portfolio pipeline with industry and style exposures, covariance estimation, exposure constraints, and attribution separating systematic exposure from stock-specific residual risk.
  • Winsorized equity inputs, estimated residual returns with pseudoinverses, and compared regularized least squares with neural networks using cross-validation for factor-realization stability.
  • Backtested on a 3,061-stock panel with 59 industry and 6 style factors (2003–2011), reviewing exposures, cumulative P&L, drawdown, attribution, and risk decomposition for portfolio review.

GitHub

Skills & Certificates

Programming

Python (3.x) SQL R MATLAB C/C++ Bloomberg Excel Jupyter Notebook Docker Git

Languages

  • English & Chinese — Bilingual
  • Spanish, French

Certificates

Interests

  • Chess — Blitz rating 2300
  • Piano, Running, Badminton