When applying, select your preferred top three home-base locations from among the Federal Reserve Banks. Each of the Banks offers its quantitative fellows unique opportunities.
For example, here’s the type of work previous fellows have done:
Atlanta:
- Conducted transaction testing models during the examinations of model risk management and business line reviews
- Developed bank-reporting tools related to risk identification and monitoring
- Experiment with ML and Generative AI using our private and public data to create unique insights
- Conducted quality control reviews on products developed by the data science team
Boston:
- Supported the National Stress Testing Program’s initiatives including development, maintenance, and risk management for the supervisory stress test models
- Participated in supervisory exams about currency risk and use of AI/ML models
- Conducted research on risks to financial stability
Chicago:
- Contributed to the Wholesale Credit Risk Center’s mission to serve as the Federal Reserve System’s leading provider of wholesale credit expertise in supervision by using advanced statistical approaches to create models and analytical tools
- Developed and maintained econometric and statistical models used in stress testing to identify and measure risks across various areas such as corporate loans, commercial real estate loans, and asset-backed securities
- Contributed to horizontal examinations and surveillance activities for wholesale credit portfolios, enhancing the oversight and evaluation of credit risks within large financial institutions
Cleveland:
- Reviewed models and technical aspects of supervisory work, such as model risk management, wholesale and credit models, and market risk
- Conducted the System’s main horizontal reviews
- Engaged with select research work on advanced analytics projects
Minneapolis:
- Assisted the System Model Validation group by validating and assessing supervisory models
- Worked with the Stress Testing Program’s Production group on the implementation and production of supervisory stress test models
- Worked with local surveillance team to develop new surveillance tools and to automate existing tools and processes
New York:
- Worked on data visualization and model development projects related to global trading and counterparty credit markets
- Performed quantitative research in areas such as stress testing, impact of financial regulations and development of liquidity risk metrics
- Developed a custom web application to support stress testing operations
- Explored applications of machine learning and natural language processing to supervisory work
Philadelphia:
- Conducted research on consumer lending risk using several of the Federal Reserve’s most robust datasets of retail lending to help identify emerging risks in the banking system
- Evaluated internal stress test models that predict how consumer loan portfolios perform during economic downturns—developed analytical tools to assess model accuracy and presented findings to senior Federal Reserve leadership
- Analyzed securities and investment portfolios to assess bank financial health during bank examinations and stress testing exercises
- Supported bank examinations and supervisory activities by conducting quantitative analysis, preparing reports, and responding to data requests from bank examination teams
Richmond (Charlotte, NC):
- Contributed to supervisory model development in various risk areas, including net revenue and wholesale credit risk, as well as modeling the effect of the global market shock
- Supported the GSIB program in multiple risk areas, including operational risk, wholesale and retail credit risk, and counterparty credit risk
- Assisted on bank examinations, focusing on model development and model risk management
San Francisco:
- Participated in the development and production of supervisory stress testing models
- Worked on data management and analytics, model development and coding, code review, and ongoing model monitoring
- Developed an expertise in market risk modeling, such as counterparty credit risk and securities, for business-as-usual and stress testing applications
- Focused on developing supervisory tools and leveraging data science techniques to enhance efficiencies in examinations and develop techniques to quantify nonfinancial risk
St. Louis:
- Utilized data analytics, predictive modeling, and visualization to enhance decision-making and operational processes
- Implemented AI, machine learning, and NLP to automate tasks, improve accuracy, and optimize workflows
- Streamlined processes through automation, workflow optimization, and effective resource management, leading to increased productivity and cost-effectiveness