Banking
AI fraud & anti-money-laundering detection
Machine learning that scores transactions for fraud and money-laundering risk and flags suspicious activity with actionable intelligence.
The challenge
Rule-based monitoring generated large volumes of alerts that analysts had to clear manually, while new fraud patterns slipped past static thresholds.
What we built
We trained gradient-boosted models (CatBoost) with NLP features on transaction data, served them through FastAPI with Swagger documentation, and integrated the results with the bank’s IBM DB2 and MySQL data stores. Each flag carries the factors behind it.
Highlights
- Real-time risk scoring of transactions
- NLP features from narrative and reference fields
- Documented API for integration with existing workflows
- Masked personal data in modelling pipelines
Outcome
Gave the compliance team risk-ranked, explainable alerts so investigation time goes to the transactions that matter most.
Free assessment
Have a process that’s slow, manual or risky? Let’s engineer a smarter one.
An engineer, not a salesperson, will reply within one business day. The first assessment is free and there is no obligation.