Lead Data Scientist - Credit & Lending

About the Company

We are collaborating with a fast-growing tech firm that innovates at the convergence of artificial intelligence, data science, and financial services. This company is creating advanced infrastructure leveraging sophisticated data and machine learning to reshape financial decision-making processes.

Position Overview

We seek a Lead Data Scientist specializing in credit risk within fintech lending or digital financial sectors. The role involves full responsibility for building and operationalizing advanced credit decision-making and risk models, emphasizing models related to probability of default and associated credit-risk frameworks.

Responsibilities

• Develop credit decisioning models by designing, validating, and deploying probability of default (PD) models and related models such as LGD, EAD, and expected loss, handling processes from raw data processing through to full production deployment.
• Investigate and engineer alternative data sources beyond traditional financial and transactional data, including behavioral and web signals, to enhance model efficacy.
• Create rigorous credit scorecards employing methodologies like Weight of Evidence (WOE), Information Value (IV) binning, monotonicity constraints, and reason codes coupled with thorough discrimination and calibration techniques.
• Manage the entire model lifecycle encompassing training, tuning, performing out-of-time and out-of-sample validations, deployment, and continuous monitoring to detect model drift and stability issues.
• Lead model governance initiatives by enforcing strict validation protocols, including champion/challenger testing, leakage identification, and comprehensive documentation.
• Serve as the principal authority on credit modeling by collaborating closely with product, engineering, and risk teams to strategically design, calibrate, and justify credit risk models.

Requirements

• Substantial experience within fintech or digital lending environments focused on credit risk.
• Proven track record in creating and deploying production credit decisioning models, with an emphasis on probability of default.
• Expertise in sourcing and utilizing alternative data sources such as behavioral and web-derived signals.
• Mastery of credit scorecard construction techniques, including WOE/IV binning, monotonic constraints, and reason code formulation.
• Robust knowledge of tabular machine learning algorithms like XGBoost, LightGBM, CatBoost, logistic regression, and ensemble tree models.
• A solid foundation in model validation practices, calibration processes, and hyperparameter optimization.
• Hands-on experience in deploying predictive models within cloud-based production frameworks.
• Comfort working with imperfect real-world datasets, with aptitude for detecting data leakage or target contamination.
• Advanced proficiency in Python programming, especially using pandas and scikit-learn libraries.
• Ability to bridge technical coding work with commercial and stakeholder communication to effectively explain, defend, and support credit risk models.

Why Join Us?

Become part of a pioneering technology company at the forefront of applying AI and data science to complex financial challenges. You'll engage with senior technical leadership and take meaningful ownership over the evolution and excellence of the firm's credit-risk modeling solutions.