IN OFFICE (CAMBRIDGE, ON) IN OFFICE (CAMBRIDGE, ON)*
IN OFFICE (CAMBRIDGE, ON) IN OFFICE (CAMBRIDGE, ON)*
Lead Credit Risk Modeler (Consumer Lending)
About the Role
We need a battle-tested model builder who's shipped
consumer lending models
that actually ran live portfolios--not a scorecard tweaker or a generic data scientist. You'll own the full cycle: designing, deploying, and tuning credit models, setting cutoffs, and shaping risk strategy with the CEO. You're as comfortable with gradient boosting as you are explaining P&L impacts to the board.
What You'll Do
Build and deploy predictive models (logistic regression, LightGBM, neural networks) for default risk and repayment likelihood using Python, SQL, or SAS.
Set risk-based cutoffs and policy overlays to maximize approvals while keeping losses low across non-prime and near-prime portfolios.
Wrangle messy credit bureau (TransUnion, Equifax), bank transaction, and internal data--no dataset is too ugly.
Monitor delinquency trends, vintage curves, and model stability (Gini, KS) to stay ahead of risks.
Justify cutoffs and strategies to leadership with clear P&L impacts (e.g., "this cutoff saves $X in charge-offs").
Recommend portfolio segmentation and risk appetite to fuel growth without blowing up.
Must-Haves
5+ years in consumer credit risk modeling for lenders (installment loans, credit cards, BNPL, auto).
Personally built and deployed at least one production consumer credit model with proven impact (e.g., cut losses by X%, boosted approvals by Y%).
Hands-on deployment of AI/ML models (e.g., LightGBM, XGBoost, neural nets) in production, not just logistic regression.
Deep knowledge of credit bureau data, tradeline behavior, and Canadian lending regulations.
Track record of setting cutoffs that balance statistical lift with real-world P&L outcomes.
What You're Not
A scorecard analyst who's only binned variables or tweaked WOE in SAS.
A model validator or compliance officer who's never shipped a production model.
An academic with theoretical PD/LGD/EAD papers but no live portfolio experience.
A data scientist who's only built Kaggle models or worked
outside lending.
Compensation
Base: $140,000-$170,000 CAD, based on experience.
Bonus: Up to 20% of base, tied to delinquency reduction and approval lift.
Location: In-office, Cambridge, ON--our models move millions, and you'll be at the heart of the business.
Why Join Us?
Join a team redefining non-prime lending in Canada. Your models will fund real people and drive our growth, not sit in a report. If you've shipped consumer credit models and think like a business operator, this is your shot to own a critical piece of our mission.
Job Types: Full-time, Permanent
Pay: $140,000.00-$170,000.00 per year
Work Location: In person
Beware of fraud agents! do not pay money to get a job
MNCJobz.com will not be responsible for any payment made to a third-party. All Terms of Use are applicable.