Resume & Background
Comprehensive overview of quantitative research leadership, predictive modeling, and statistical systems.
Leilei Zhou, Ph.D.
Quantitative Researcher | Systematic Research | Statistical Modeling
Ph.D.-trained quantitative researcher with 8 years of experience building, validating, and productionizing statistical and machine-learning models at J.P. Morgan. Brings 4 years of quantitative team leadership and deep experience in time-series analysis, factor and signal research, credit and macro modeling, robust backtesting, and scalable research infrastructure.
Work Experience
Lead Quantitative Researcher
Lead a 2-person quantitative research team through end-to-end development, mathematical validation, and production deployment of predictive models supporting billions of dollars in institutional capital.
- Lead a 2-person quantitative research team through end-to-end development, mathematical validation, and production deployment of predictive models supporting billions of dollars in institutional capital.
- Develop state-space and recursive Bayesian models using Kalman filtering to track time-varying relationships and latent states across high-dimensional market data.
- Build PCA-based factor extraction workflows across correlated macroeconomic clusters and apply stationarity, autocorrelation, heteroscedasticity, and ADF diagnostics to isolate orthogonal predictive factors.
- Model loss rates with regime-switching models and structural-break tests to identify changes in market behavior and recalibrate state-transition parameters.
- Develop distance-to-default and corporate credit models using CDX HY spreads and U.S. Treasury yields; model duration and convexity to capture cross-asset sensitivities.
- Engineer Spark, DuckDB, and vectorized NumPy data pipelines that process 20+ years of panel data across 50+ million facilities; resolve out-of-core constraints with lazy evaluation and targeted column selection.
- Architect production deployment workflows from Git staging through live engines, applying CI/CD, Docker, and pytest-based test coverage to improve reproducibility and release discipline.
- Created a Python library that parallelizes 12 complex statistical models, eliminating manual interventions and saving 50 compute and review hours per run.
- Integrate LLM and MCP-enabled tooling to query databases, accelerate forecast-output extraction, refactor legacy code, and generate rigorous test suites while maintaining production coding standards.
Quantitative Science Intern
Researched multiple-testing controls and engineered scenario-analysis tools for statistical inference.
- Developed multiple-hypothesis testing frameworks in R and used Monte Carlo simulations to quantify and control false discovery rates.
- Built an interactive scenario-analysis application that accelerated evaluation of complex statistical-testing strategies by 30%.
Education
Ph.D. in Statistics
Stony Brook University
Relevant Coursework: Machine Learning, Advanced Time Series Analysis, Multivariate Analysis, Simulation