Data Scientist

Job description

Key Responsibility Area : (Specifies Key Result Areas for the Incumbent)

o Apply predictive modeling techniques, including decision trees, regression analysis, and machine learning,

to develop models that meet predictive accuracy requirements.

o Evaluate the conceptual and technical soundness of valuation and risk models according to OCC SR 11-7

Model Risk Management guidelines.

o Critically assess modeling approaches, frameworks, assumptions, and testing methodologies, covering

performance testing, back testing, benchmarking, and stress / sensitivity testing.

o Perform data extraction from relational databases using SQL and employ programming languages like

Python and R for model testing, data analysis, and generating reports.

o Ensure model documentation adheres to regulatory standards, particularly SR 11-7 guidelines, covering

data analysis, model methodology, performance evaluation, and implementation.

o Execute risk analysis to validate the performance of existing risk models, lead gap-closing initiatives, and

recommend improvements.

o Independently review model validations, assess identified issues, and propose solutions that meet

methodological rigor.

o Generate regular reports on model performance monitoring, including updates on compliance with policies

and procedures, progress, challenges, and issues.

o Provide recommendations for model design, development, back testing, implementation, and recalibration

processes.

o Assist in model performance reviews, validate controls, and ensure effective communication with

stakeholders about validation results.

o Develop and maintain a comprehensive and accurate model inventory in alignment with Model Validation

Policy requirements.

Eligibility Criteria : (Skill set required to do the job- Knowledge, Tools, Technical Knowledge, Certifications etc.)

o Proficiency in predictive modeling techniques, including decision trees, regression analysis, and machine

learning.

o Strong programming skills, especially in SQL, Python, and R, for model testing and data analysis.

o Proficiency in machine learning libraries (e.g., scikit-learn, TensorFlow, PyTorch) and also experience with

big data platforms (e.g., Spark, Hadoop).

o Familiarity with cloud platforms (e.g., AWS, GCP, Azure) for data storage and model deployment.

o Thorough understanding of model risk management and regulatory standards, particularly OCC SR 11-7

guidelines.

o Expertise in model documentation, including methodologies, data analysis, and performance assessment.

o Experience in risk analysis and gap-closing initiatives for model validation.

o Competence in performing model performance testing, including back testing, benchmarking, and

stress / sensitivity testing.

o Ability to effectively communicate validation results and model performance insights to stakeholders.