AML Data Analyst: Modeling & Tuning (Python/SQL)

Matrix Usa Inc; United States; On-site; USD 90,000 - 130,000; Full time

Job summary

Matrix seeks an experienced Data Analyst with a strong banking background and expertise in financial crime analytics. You will develop and tune classification models (logistic regression, multinomial logit, XGBoost, LightGBM, RF) and perform data analysis using Python and SQL.

The role focuses on sanctions, fraud, and AML use cases, evaluating model performance, and presenting findings to senior management and regulators. Collaboration across functions is essential.

Qualifications

• Master's or PhD in Statistics, Economics, Finance, Mathematics, Data Science or related quantitative field.
• Experience in banking or financial services, sanctions, fraud, or AML modeling preferred.
• Proficient in Python and SQL with production-quality coding ability.

Responsibilities

• Perform quantitative analysis and model tuning for sanctions, fraud, and AML use cases.
• Develop, validate, and optimize statistical and ML models for detection accuracy.
• Apply tuning methodologies to identify opportunities for model optimization.
• Develop and evaluate classification models including logistic regression, multinomial logit, and tree-based models (XGBoost, LightGBM, RF).
• Use Python and SQL for data analysis, feature engineering, modeling, and automation.
• Work with large datasets to identify data quality issues and modeling gaps.
• Compare model performance using statistical measures and recommend optimizations.
• Document tuning methodologies, results, and metrics.
• Prepare reports and present findings to senior management and regulators.
• Collaborate across business, technology, compliance, and governance teams.
• Maintain code quality and version control (GitHub).

Skills

Python

SQL

Statistical modeling

Machine learning

Data analysis

Model development

Education

Master's or PhD in Statistics or related

Tools

GitHub

Job description

Matrix seeks an experienced Data Analyst with a strong banking background and expertise in financial crime analytics. You will develop and tune classification models (logistic regression, multinomial logit, XGBoost, LightGBM, RF) and perform data analysis using Python and SQL.

The role focuses on sanctions, fraud, and AML use cases, evaluating model performance, and presenting findings to senior management and regulators. Collaboration across functions is essential.