Yiqing(Lucy) Wang

Yiqing(Lucy) Wang

About me

I am currently a Model Risk Validator at Citigroup, specializing in Retail Risk & Consumer Valudation Models. In this role, I evaluate the soundness, regulatory compliance, and business relevance of Statistical Models and Machine Learning Models used in both customer acquisition and existing customer management strategies.

I hold a Ph.D. in Statistics, with research focuse on applying deep learning to biomedical challenegs- specifically, predicting antimicrobial resistance at the genetic level using advanced deep learning algorithms.

My expertise spans model validation, machine learning, statistical modeling, and data analysis. I am passionate about applying analytical methods to real-world problems, and I’m especially interested in advancing model risk governance, quantitative risk management, and AI-driven decision frameworks in financial services.

I am seeking opportunities in Quantitative Risk or Model Development, particularly in Fixed Income, Credit Risk, or Market Risk domains.

Working Experience

  • Model Risk Management Validator Citigroup, Dallas TX

    • Lead validation of consumer credit risk models (e.g., for acquisition and account management), ensuring alignment with SR11-7 and Basel regulatory requirements
    • Deliver effective challenges to model assumptions, feature engineering, and variable reduction techniques across statistical and ML models (e.g., logistic regression, XgBoost)
    • Collaborate with model developers, sponsors, and business teams to communicate model limitations, risk exposure, and compliance implications. * Spearhead research initiatives on SAS-to-Python code migration, Confirming Model Development, and Integration of ChatGPT into validation processes.

  • Research Assistant UT Southwestern Medical Center, Dallas TX

    • Led a 3-person team in planning and executing data-driven research projects, supporting five studies involving large-scale real-world datasets.
    • Built scalable ETL pipelines for cleaning, transforming, and engineering features from ~1 billion Electronic Health Records using Pandas and PySpark.

Education