Remote-First Bank Model Risk Analytics Lead

Affirm; Denver (CO); On-site; USD 195,000 - 280,000; Full time

Benefits offered by this job

Equity

Remote-first culture

Health coverage

Job summary

Affirm is seeking an experienced Bank Model Risk Management professional to lead validations of sophisticated credit and fraud models and to build automated monitoring in Python. You will partner with model developers and control functions to ensure models are robust and compliant.

The role emphasizes deep quantitative skills, SQL expertise, and effective communication to translate complex findings for diverse audiences.

Qualifications

• 7+ years of experience in technical credit/fraud risk modeling, model validation or quantitative analytics.
• Deep understanding of consumer credit lifecycle and fraud detection.
• Expertise in loss forecasting, fraud prediction and stress-testing frameworks.
• Proficient in Python with pandas, scikit-learn and statsmodels, and SQL for data wrangling.

Responsibilities

• Full-Stack Model Validation of credit/fraud models, focusing on soundness, data integrity and performance stability.
• Develop automated monitoring suites in Python to track drift and PSI in real-time.
• Remediate validation findings in collaboration with 1st-line model developers.
• Partner with Audit, Controls and Compliance for timely regulatory requests.
• Support the Bank Model Risk Management function for bank-owned models.

Skills

Python

SQL

Model validation

Credit risk modeling

Fraud modeling

Data analysis

Communication skills

Tools

Pandas

scikit-learn

statsmodels

Job description

Affirm is seeking an experienced Bank Model Risk Management professional to lead validations of sophisticated credit and fraud models and to build automated monitoring in Python. You will partner with model developers and control functions to ensure models are robust and compliant.

The role emphasizes deep quantitative skills, SQL expertise, and effective communication to translate complex findings for diverse audiences.