ML Validator, Model Risk and Analytics Division

The responsibilities of this specialist include the validation, monitoring, and improvement of machine learning models across the entire bank, ensuring data quality, model reliability, explainability, and compliance with model risk management standards.

Development and provision of data quality control measures for training and validating machine learning models in various bank business processes

Development of methods and principles for the validation and testing of machine learning models in the bank

Development of automated validation systems for machine learning microservices

Development and provision of the model risk management process

Development of monitoring systems for machine learning model quality

Improvement of business analytics and model validation methodology

Development of technical requirements for model validation components

Reporting and informing stakeholders about model validation and quality control (explainability and interpretability)

Recommendations for improving the machine learning model lifecycle system

Execution of other assignments from the supervisor

Higher education in the field of statistics, mathematical modeling, mathematics, risk management, or engineering

Deep knowledge of statistics and probability theory, knowledge in the field of financial analysis and machine learning models

Understanding of interpretability analysis of statistical models (regression analysis, decision trees, gradient boosting, etc.) and deep learning models

Experience working with SHAP and ELI5 schemes will be considered an advantage

Understanding of problems related to data, model, and concept drift, as well as ways to resolve them

Development experience in Python, including Poetry, Git, code checking tools, pip, and Docker

Analytical and creative thinking

Communication skills and ability to work in a team; interaction and coordination skills

Excellent knowledge of Armenian and English languages