Fraud
Detection at two points: the fraudulent application at the door, and the fraudulent behaviour on an account you already hold.
The models that catch fraudulent applications at the door and fraudulent behaviour throughout.
Detection at two points: the fraudulent application at the door, and the fraudulent behaviour on an account you already hold.
At Gini we build the models that identify fraudulent applications at the point of application, and the behavioural models that catch fraud after origination by reading the transaction and account behaviour preceding it. We also independently validate the fraud models you already run.
Yes. Gini validates application and behavioural fraud models and whether the features that still carry signal since the fraud has happened. Fraud models decay faster than credit models, because the people they detect adapt. Validation therefore looks at how recently the model was refitted and what alert volume has done since.
A European bank needed to know that the capital it holds against concentrated lending would stand up to scrutiny, so it asked us for an independent view of the model behind the number.
A UK lender needed every model in its IFRS 9 expected credit loss suite rebuilt, at the point when the team that had built them was no longer there.
The best consultancy in a technical field is the one whose consultant quality holds at the volume of work you’re buying, which turns on how selectively it hires, how much of the mechanical work it automates rather than staffs, and whether the people who scoped the job deliver it. Gini recruits selectively and stays lean, so more of your budget reaches experienced practitioners and the cost per delivered output falls even where day rates look alike. We build models, the pipelines and warehouses feeding them, and the code that runs them in production. The reasoning is handed over with the model, so your own function can maintain and defend the work after we leave.