Quant Modeling Lead [multiple Positions Available]

JPMorgan Chase JPMorgan Chase · Banking · Jersey City, NJ +1 · Corporate Sector

Lead Quant Modeler responsible for independent testing, benchmark model development, and evaluating risk associated with advanced ML/DL models, particularly in fraud detection. This role involves guiding model developers, ensuring adherence to risk management and regulatory requirements, and overseeing the end-to-end model risk lifecycle. The position also includes mentoring junior team members and representing the team in regulatory reviews.

What you'd actually do

  1. Conduct independent testing and develop benchmark models to serve as reference points for model performance.
  2. Evaluate risk associated with advanced machine learning and deep learning models by identifying potential vulnerabilities and recommending mitigation strategies to minimize model risk.
  3. Assess and challenge model design and implementation to ensure alignment with risk management and regulatory requirements.
  4. Oversee the end-to-end model risk lifecycle, including validation, performance monitoring, change management, and issue remediation.
  5. Represent the team in audit and regulatory reviews, ensuring timely and accurate responses to inquiries.

Skills

Required

  • Model risk assessments of XGBoost models
  • Validation of statistical methods, hyper-parameters, and interpretability using Shap
  • Independent reviews of ML pipelines
  • Validating feature engineering, selection processes, tuning methodologies, and evaluation frameworks using XGBoost, Scikit-learn, PyTorch, TensorFlow, and Keras
  • Benchmarking and analyzing data using algorithms including linear and logistic regression, clustering, and CART
  • Evaluating models using metrics including AUC-ROC, Precision-Recall, cross-entropy loss, and KS
  • Developing models using Python, R, and SQL with object-oriented programming principles
  • Manipulating data including imputation, encoding, and normalization using NumPy, Pandas, PySpark, and Dask
  • Visualizing results with Matplotlib, Seaborn, and ggplot
  • Optimizing dataset computations with multithreading and multiprocessing in PySpark
  • Managing data storage and processing with AWS
  • Conducting analysis in graphical database Using TigerGraph

Nice to have

  • Mentor and train junior team members
  • Draft and review documents related to model review
  • Coordinate and lead governance activities for the team such as performance monitoring and the annual status assessment
  • Provide insights and recommendations for model enhancements
  • Present findings to senior management and key stakeholders
  • Support team development initiatives, including recruiting, hiring, and onboarding new team members

What the JD emphasized

  • model risk governance
  • regulatory requirements
  • independent testing
  • advanced machine learning and deep learning models

Other signals

  • model risk governance
  • independent testing
  • benchmark models
  • advanced machine learning and deep learning models
  • fraud detection techniques
  • model validation process
  • model risk lifecycle
  • audit and regulatory reviews