Like all banks, one of Belgiumās biggest banks is constantly faced with new fraud threats whereby criminals are trying to steal from their unsuspecting clients.

Challenge
Achieving fraud detection is only possible by executing real-time analysis, preventing the transaction instead of having to make corrections after the fact, when the damage could already be done. This ensures that the bank and its clientele are protected against fraudulent transactions.
Solution
The European enforcement for banks to process transactions within 1 day has had a great impact on banksā infrastructure. The requirement for banks to provide instant payments to their customers whereby payments are processed in a couple of seconds leaves fraud detection pipelines with 350ms to āscoreā (deciding whether the transaction is fraudulent or not) each transaction. Klarrio created a high-level Domain-Specific Language (DSL) workbench using Apache Flink to enable the bankās data scientists to more easily interact with the fraud detection application. In addition, Klarrio provided the data scientists with a local environment to develop, test, and analyze data. As a result, the implementation provided the bank with a Continuous Integration / Continuous Deployment (CI/CD) environment.
The Expertise
- Real-time analysis
- Continuous Integration
- Continuous Deployment
The Technology behind
- Flink
- CI/CD
- Domain-Specific Languages
”āThings work out best for those who make the best of how things work out.ā
ā John Wooden
Results
Real-time | maintaining the SLA of fraud detection within 350ms of payment.
Ease of Use | provide bankās data scientists with advanced capabilities without having to deal with low-level code.
Ask for a demo!
We’re your one-stop cloud-native partner
Big Data engineering. Data science. Data Analytics. Site Reliability Engineering. Consulting. And customized Open Source projects for companies of all sizes. Learn more about what Klarrio can do for you today.
Other Projects
Just a few projects examples.