Strategic Wholesale Credit Risk Modeling VP

Testing Site - JPMC; Jersey City (NJ); On-site; USD 200,000 - 260,000; Full time

Job summary

JPMorgan Chase is seeking a Risk Management - Wholesale Credit Risk VP to lead model development, design advanced forecasting models, and support CCAR, ICAAP, and CECL initiatives. The role requires strong statistical methods knowledge, Python programming, and experience with credit risk at both single-obligor and portfolio levels.

Responsibilities include building ML-enabled risk models, data processing, and ongoing performance evaluation, with collaboration across model, data, and

Qualifications

• Solid theoretical and practical knowledge of statistical methods and models (GLM, time-series, clustering, logistic regression).
• Hands-on Python programming experience and data cleaning/filtering of large datasets.
• Experience with credit risk modeling at single-obligor and portfolio levels.
• Experience writing documents for regulatory reviews.

Responsibilities

• Lead model development: evaluate forecasting models and risk engines for Wholesale Credit.
• Design, enhance, and maintain statistical and econometric models for forecasting and risk management.
• Develop Stress Capital modeling for CCAR/ICAAP/CECL programs.
• Utilize and implement AI/ML techniques for risk monitoring and large datasets.
• Collaborate with data and model implementation teams; monitor model performance regularly.

Skills

Statistical modeling

Python

Time-series

GLM

Machine Learning

Data cleaning

Credit risk

Regulatory documentation

Tools

R

Job description

JPMorgan Chase is seeking a Risk Management - Wholesale Credit Risk VP to lead model development, design advanced forecasting models, and support CCAR, ICAAP, and CECL initiatives. The role requires strong statistical methods knowledge, Python programming, and experience with credit risk at both single-obligor and portfolio levels.

Responsibilities include building ML-enabled risk models, data processing, and ongoing performance evaluation, with collaboration across model, data, and