I am the founder and CEO of Traustia, a company building independent validation for biomedical prediction claims, and a Doctor of Technology (D.Tech.) candidate at Purdue University. I hold an A.L.M. in Data Science from Harvard Extension School, an M.S. in Mathematics and an M.A. in Mathematics Education from California State University, Los Angeles, and graduate preparation in financial engineering.
My research is on a single question: how much confidence a quantitative result actually deserves. It centers on leakage-aware validation — the methods that keep a model from being scored on information it should never have seen — together with uncertainty quantification, calibration, and reproducible analytical workflows.
Much of this is inherited from quantitative finance, a field that learned early and expensively that a backtest is not a result. Walk-forward validation and strict timing rules exist there because it is remarkably easy to build a model that predicts the past. Biomedicine is now confronting the same failure under a different name: data leakage.
I have taught mathematics, statistics, data science, and information security at California State University, Los Angeles since 2012. I care about work that is transparent, rigorous, and genuinely useful.
Two student research teams I led and co-advised are bringing reproducibility, evidence auditing, and biomedical data science into international research forums.
From Shared Cancer Data to Auditable Evidence: A FAIR Audit of Breast-Cancer pCR Prediction
I led a student team auditing whether shared breast-cancer data can support transparent, reproducible pCR prediction. The work was accepted by the NCI Office of Data Sharing's Fourth Annual Data Sharing Symposium and received a travel award.
View the NCI event ↗Evidence Mapping of OPM-MEG and SQUID-MEG Readiness for Mild Traumatic Brain Injury
I co-advised a student research team comparing OPM-MEG and SQUID-MEG readiness for mild traumatic brain injury through evidence mapping. The paper was accepted for presentation at the 2026 IEEE MIT Undergraduate Research Technology Conference.
View the MIT URTC event ↗Where the research lives — from validation methods to biomedical prediction research and decision-oriented risk analytics.
Leakage-aware validation and reproducible computational methods.
Calibration, VaR & Expected Shortfall, and tail-risk estimation.
Research on biomedical prediction claims, public omics data, and leakage-aware evaluation.
M.S. Mathematics · M.A. Education
A.L.M. in Data Science · Extension School
Doctor of Technology (D.Tech.) · In progress
Trustworthy machine learning, biomedical evidence, and decision-oriented risk — united by leakage-aware validation, reproducibility, and clear communication of uncertainty.
Leakage-aware validation, secure data workflows, responsible AI evaluation, uncertainty communication, and simulation-based validation.
Research on public omics classification, dataset integrity, leakage-aware evaluation, and the limits of translational claims.
Financial and operational risk, calibration assessment, VaR and Expected Shortfall, EVT/GPD tail-risk, and volatility-managed decision systems.
Cybersecurity risk frameworks (NIST CSF, AI RMF, OWASP), AI-enabled threats, security analytics, and algorithmic risk mitigation.
Linking public regulatory records for emissions and vehicle-safety early-warning modeling under strict temporal validation.
Curriculum design, quantitative reasoning, and pedagogy that makes statistical thinking transparent and durable.
Adjunct Instructor, Mathematics · Statistics · Data Science · Cybersecurity · California State University, Los Angeles · 2012 – present
I teach undergraduate courses spanning statistics, regression, introductory data science, quantitative reasoning, information security, and security risk management. I design applied assignments and assessments that connect data interpretation, quantitative modeling, cybersecurity concepts, risk awareness, and responsible technology use — and mentor students in reproducible analysis habits and security-aware pathways.
I also coordinate the 2026 Summer Biomedical Engineering & Quantum Technology Program at UCLA, supporting interdisciplinary research preparation across biomedical engineering, emerging technology, and quantitative reasoning.
Earlier in my career, I taught mathematics at Tianmu Junior High School in Taipei, Taiwan.
A selection of public data-science and machine-learning projects — spanning causal inference, computer vision, finance, and NLP.
Causal impact evaluation of an education program using a difference-in-differences design.
Extracting market sentiment signals from text, in a reproducible notebook workflow.
A 2025 summer project exploring intraday trading signals and execution logic.
Customer-journey analytics on experiential-venue data, in a reproducible notebook.
Vision-based navigation for a Duckiebot on NVIDIA Jetson Nano, combining sensors and computer vision.
For paper collaborations, research, teaching, or speaking, please use the contact address below.
For validation work and company enquiries, please contact Traustia.