AI Default Prediction Specialist
An AI Default Prediction Specialist designs, trains, and operationalizes machine-learning models that forecast the probability of …
Skill Guide
The combined understanding and application of international (Basel III, IFRS 9) and US (CECL) regulatory capital and accounting frameworks, focusing on how they dictate the construction, validation, and use of credit risk models for capital adequacy, expected credit loss provisioning, and financial reporting.
Scenario
You are given a dataset of 100 corporate loans with vintage, industry, and financial ratios. You must calculate a 12-month ECL under IFRS 9 Stage 1.
Scenario
Your bank's consumer portfolio (credit cards, personal loans) shows rising delinquencies due to an economic downturn. You must update the staging model to correctly identify borrowers with a Significant Increase in Credit Risk (SICR).
Scenario
The CFO and CRO are at odds: Basel III RWA models are producing one risk estimate, while IFRS 9 ECL models are producing another, causing confusion at the board level. You must build a reconciliation and stress-testing framework.
Core for developing, validating, and implementing credit risk models (PD, LGD, EAD). Python is increasingly dominant for its flexibility in handling data pipelines and machine learning techniques within these structured frameworks.
Non-negotiable for documentation, validation, and audit defense. SR 11-7 defines the standards for model development and independent validation. Knowledge of these is what separates a modeler from a model risk manager.
For sourcing macroeconomic data (GDP, unemployment, housing prices) critical for forward-looking models, and for benchmarking against market-implied parameters (e.g., bond spreads for PiT calibration).
Answer Strategy
Demonstrate a structured, root-cause analysis approach. **Sample Answer:** 'First, I'd decompose the ECL change: isolate the impact of updated macroeconomic forecasts from changes in portfolio composition or model re-calibration. Second, I'd review the model's sensitivity to the key macro driver-likely unemployment-and check if the input scenario is still within the range of historical experience. Third, I'd present the findings to the business head with data: showing the provision is driven by a specific, justifiable risk factor, not model error, and discuss if the portfolio's risk profile has genuinely changed.'
Answer Strategy
Test the candidate's ability to articulate regulatory intent and design efficient systems. **Sample Answer:** 'Basel III focuses on *capital adequacy* for unexpected losses, using a standardized or IRB approach with prescribed downturn parameters. IFRS 9 focuses on *profit and loss provisioning* for expected losses, requiring forward-looking, point-in-time estimates. To serve both, I'd build a core PD/LGD/EAD engine with a common data foundation, then layer on a 'regulatory parameter adapter'-this module would apply downturn adjustments for Basel and forward-looking scenarios for IFRS 9, ensuring consistency in the base risk estimates while satisfying divergent regulatory objectives.'
1 career found
Try a different search term.