Director of Credit & Analytics

Billd•Austin, TX

1d•Hybrid

About The Position

The Director of Credit and Analytics will lead credit risk modeling and portfolio analytics at Billd, reporting directly to the Chief Operating Officer. The role is responsible for designing, developing, and overseeing decisioning models for underwriting, credit limit assignment, and risk-based pricing tiers across Material Financing, Pay App Advance, and Flex Line. Billd’s models evaluate subcontractor, project, and general contractor risk together to forecast subcontractor cash flow and payment timing through the commercial construction payment chain. The framework is built on CHAID decision trees and may use regression and gradient boosting (XGBoost) to improve predictive performance. This senior, hands-on role is approximately 90% analytical. The Director sets the strategic direction for credit models and performs quantitative work across feature engineering, model development, validation, and performance monitoring, integrating AI-assisted analytics into daily work. The role also owns credit reporting to the CEO, board of directors, and capital partners. It partners with Operations and Underwriting, which apply models to individual transactions; the Director does not process or approve deals.

Requirements

8+ years of experience in B2B credit, such as commercial credit, small business finance, B2B fintech, or working capital.
4+ years of hands-on advanced credit modeling experience, including ownership of production credit risk models.
Experience in CHAID decision-tree modeling.
Expertise in statistics and predictive modeling, including regression, gradient boosting (XGBoost), and explainability methods such as SHAP.
Experience with probability of default, vintage and roll-rate analysis, survival models, backtesting, and performance monitoring using measures such as KS, Gini/AUC, and PSI.
Advanced Python and SQL skills.
Daily, hands-on use of AI tools such as Claude or Codex for coding, data analysis, and modeling, including validation of AI-generated output.
Bachelor’s degree in a quantitative field.
Ability to communicate complex quantitative findings clearly to a board, executive team, or capital partners.
Sound judgment in interpreting model results in the context of the business and credit cycle, and ability to work independently and own projects from start to finish.

Nice To Haves

Familiarity with Salesforce and BI tools such as Metabase.
Experience underwriting with bank transaction data; strongly preferred.
Experience in construction, receivables, factoring, or trade credit.
Master’s degree or PhD in statistics, mathematics, economics, data science, or a related field; strongly preferred.

Responsibilities

Lead the design, development, and governance of CHAID-based decisioning models, using regression and gradient boosting (XGBoost) where they improve predictive performance.
Manage the model lifecycle, including development, validation, deployment, monitoring, and recalibration.
Model contractor, project, and general contractor risk to forecast subcontractor cash flow and payment timing across the construction payment cycle.
Engineer predictive attributes from bank transaction data, commercial and consumer credit bureau data, lien and judgment records, project data, and internal repayment history; assess third-party data sources for incremental predictive value.
Translate model output into credit tiers, credit limit methodology, and product eligibility for Material Financing, Pay App Advance, and Flex Line; recommend calibrated changes to credit policy thresholds.
Define the credit modeling roadmap and assess trade-offs among growth, approval rates, and loss performance.
Own credit reporting to the CEO, board of directors, and capital partners, including portfolio performance and concentration, delinquency and loss trends, vintage and roll-rate analysis, and model performance.
Present model changes and recommendations to the Credit Committee.
Monitor model discrimination, calibration, and stability over time.
Lead loss forecasting, scenario analysis, and stress testing, and identify early indicators of portfolio deterioration.
Integrate AI tools into model development, coding, and analysis, while maintaining rigorous, reproducible, and auditable documentation.
Partner with Operations, Underwriting, Finance, Engineering, and Product to implement model changes in production decisioning systems.