After nearly two decades in the payments space, Arun Kumar is once again redefining how the industry thinks about payments. Known for his unique ability to connect technology and business in challenging times, Kumar has become the trusted name across global banks and fintechs.
Over his 19-years of career, he has left his mark across continents, be it helping Ingenico (USA POS manufacturer) enable vital EBT transactions in their POS machines, or advising Mastercard on its acquisition of the ElectraCard Payment Product from his Company, and assisting Mastercard launch their first payment-processing platform using ElectraCard Suite of Products in the Middle East. His footprint extends to modernizing the banking product and technology infrastructure of ACLEDA Bank in Cambodia, Bramer Bank in Mauritius, and Bank of Baroda in India. During his career he also built and launched critical compliance and risk-management products and solutions at Western Union under the guidance of the U.S. Department of Justice (DOJ) and the Federal Trade Commission (FTC). At Western Union, he also built advanced KYC solutions that significantly improved global ID capture and verification efficiency.
Arun has been driving the payment industry with innovations. But what’s new? This time he has been doing the transformations through Artificial Intelligence to test and validate ISO 20022 messages faster and more effectively.
As he says, “necessity is the mother of innovation”. Between mid-2023 and early 2025, while at Cognizant, Arun was supporting several major U.S. banking clients through a high-pressure modernization program to adopt ISO 20022 for both Fedwire and SWIFT payments. To shed some light on Fedwire and SWIFT, the Fedwire is the real-time system that moves money between U.S. banks. SWIFT is the global messaging network banks use to instruct each other on payments across countries. So what really happened? Well, the clock was ticking, Fedwire’s migration from FAIM (older messaging standards) to ISO 20022 was scheduled for July 2025, and SWIFT’s transition from MT (older messaging standards) to ISO 20022 is about to follow in November 2025. Arun was required to help the banking clients with the full technical, functional, and operational readiness within tight deadlines.
Arun’s challenges were monumental. Banks faced months of manual work to create and test thousands of ISO 20022 messages required for OFAC sanctions and fraud-screening validation. That is when Arun thought differently, and he proposed leveraging GenAI to automate, and bulk generate the test cases and test data to reduce the manual work. He worked with the Cognizant’s AI engineers and designed a Gen-AI based tool that transforms a few targeted prompts with test data and test file samples into extensive batches of realistic ISO 20022 messages for both Fedwire and SWIFT formats.
The GenAI based tool produces valid and invalid messages, edge-case scenarios, and fraud-like patterns. The tool also supported direct integration into payment and screening APIs while maintaining full traceability from prompt to payload for audit and compliance teams. The impact was immediate and drastic i.e. the testing cycles shrank dramatically, coverage and data quality increased. That’s when the banks gained confidence in meeting ISO 20022 milestones ahead of schedule. This innovative approach reduced cost and risk drastically while improving release quality and timelines. How:
● Regression testing compressed from months to days.
● Load and stress testing reached near production-level realism.
● Boundary checks became easy to repeat, helping teams identify issues in fraud-screening systems and models faster.
● As issues were found faster, the right corrections were also made faster and hence made the Fraud screening systems and models more robust.
Arun ensured every generated dataset was fully governed and auditable, empowering compliance teams to trace exactly how each message was created. Arun believes this approach can extend far beyond ISO 20022 the transformations or testing projects. This Gen-AI tool and method can be applied to any scenarios where the large-scale transaction data set generation is needed for stress testing purposes, and it could be applied to any messaging standards like ISO8583, ACH, or even proprietary formats. This matters because, in modern payments or Fintech space, newer payments solutions or enhancements need to be continuously launched, and that too often within tighter regulatory deadlines. Traditional manual test-data creation and testing are slow and limited in scope. By productizing test-data generation, Arun Kumar’s AI-driven solution flips that equation—accelerating delivery while simultaneously strengthening the compliance controls. This new innovative approach finally helped all his banking clients successfully complete multiple rounds of the ISO 20022 testing way before the deadlines.
Blending engineering brilliance with an instinct for solving real-world challenges, Arun Kumar has carved a unique place for himself in the global payments industry. His work with banks and fintech firms across continents proves that innovation isn’t confined to laboratories, it thrives wherever technology has to meet human need.
As the payments industry steps into the ISO 20022 era, Arun Kumar stands as an example that the future belongs to those who dare to innovate even before the world demands. And once again, he proved that when it comes to the future of finance, Arun Kumar remains a few steps ahead of the curve.
