Data Scientist - Mortgage Financing Modeling

U.S. Citizens or U.S. Permanent Residents will be given priority consideration

Associate Level: 0-2 Years of Experience

Senior Level: 3-5 Years of Experience

We are seeking Data Scientists at both the Associate and Senior levels to support a client project in the mortgage financing industry. The role will focus on the development, testing, implementation, and documentation of statistical models and analytical applications that support business and risk decisions.

This position is part of a modeling team responsible for developing robust, scalable, and interpretable models that support portfolio risk management, loss forecasting, and other business modeling needs.

• Develop analytical methods and statistical models related to mortgage portfolio collateral, risk management, loss forecasting, and related business needs.
• Provide innovative, detailed, and practical solutions to a range of demanding and complex analytical problems.
• Implement statistical models using efficient software languages.
• Code model prototypes, prepare specifications and test cases, and modify source code in existing applications.
• Coordinate model testing through the implementation process.
• Conduct back-testing to monitor model performance.
• Perform economic tests and stress tests to validate model forecast results.
• Provide modeling and analytical support to a line of business or product area as a day-to-day technical specialist.
• Prepare technical documentation, model development rationale, and analytical support materials to comply with model oversight and model review requirements.
• For the Senior level, work under limited direction and independently develop approaches to solutions.

• Master's degree required in quantitative finance, statistics, economics, mathematics, data science, computer science, or a related quantitative field. Ph.D. preferred.
• Coursework or work experience applying predictive modeling techniques from finance, statistics, mathematics, data science, or computer programming to large data sets.
• Relevant coursework may include statistics, mathematical programming, optimization, machine learning, computational methods, design and analysis of algorithms, Bayesian methods, derivatives, or Monte Carlo methods.
• Coursework or work experience writing statistical or optimization programs to develop models and algorithms.
• Programming experience with one or more of the following: SAS, Python, R, SQL, MATLAB, or similar tools.
• Experience working with large data sets and relational databases.
• Strong quantitative, analytical, programming, and communication skills.

• 0-2 years of relevant experience beyond a Master's degree. Ph.D. preferred.
• Strong academic background in quantitative modeling, statistics, data science, finance, economics, mathematics, or related fields.
• Prior internship, academic project, research, or coursework experience involving predictive modeling, statistical programming, or large data sets is preferred.

• 3-5 years of relevant experience beyond a Master's degree. Ph.D. preferred.
• At least 3 years of experience in model testing and/or model development.
• Doctorate degree preferred.
• Experience independently developing, testing, implementing, and documenting statistical models is preferred.
• Experience supporting model review, model oversight, or model validation processes is preferred.

• Experience in statistical model development and implementation.
• Experience with software development and system setup for model applications.
• Experience in mortgage finance, financial services, credit risk, portfolio risk, loss forecasting, or related analytical domains.
• Ability to develop practical, well-documented, and interpretable modeling solutions.
• Ability to work effectively with business, modeling, technology, and model oversight teams.

• Exceptional quantitative and analytical skills.
• Strong knowledge of statistical models, tools, and techniques.
• Strong programming skills.
• Strong communication skills.
• Ability to solve complex problems with practical, business-oriented modeling approaches.