Falcon has launched MerlinFX, a suite of seven AI agents designed to carry out banking operations within its processing core.
The AI agents are built to handle functions including reconciliation, disputes, risk monitoring, customer service, campaign management, premium-cardholder engagement and card inventory. They operate on a bank’s ledger and follow its rules, while actions that cannot be reversed require approval from a designated bank officer.
Two of the agents, ReconFX and SurgeFX, are already in production with Falcon’s CardFX and CreditFX stacks at banks including Karur Vysya Bank and Jana Small Finance Bank. The remaining five agents are being rolled out among clients. Falcon’s platform processes more than 15 million cards and accounts.
Unlike AI tools that analyse information exported from a bank’s core system, MerlinFX operates directly within Falcon’s processing core and real-time infrastructure. Its agents can respond as transactions occur, with SurgeFX able to act when a transaction is authorised and ReconFX matching settlement files against the ledger as they arrive.
In an initial pilot, ReconFX auto-matched the majority of transactions and identified breaks and leakage that had previously gone unnoticed.
The MerlinFX suite includes SurgeFX for customer segmentation, campaign testing and offer launches; ReconFX for transaction matching and prioritisation of breaks; RiskFX for monitoring repayment, utilisation and transaction behaviour; and DisputeFX for handling chargebacks, UPI reversals and failed transactions.
ResolveFX manages routine customer requests through WhatsApp, SMS and voice in the customer’s own language, while ConciergeFX manages proactive services for premium cardholders. OpsFX forecasts branch-level card demand and reallocates available kits before new orders are placed.
Existing Falcon clients can activate the agents through configuration, while institutions using other core systems can connect with MerlinFX through Falcon’s APIs.
MerlinFX operates within Falcon’s security and compliance framework, which is certified to PCI DSS, PCI 3DS, PCI PIN and ISO/IEC 27001 and undergoes annual CERT-In empanelled testing. The platform can be deployed on the cloud or on-premises.
Actions involving money movement, credit-limit or account-status changes, or commitments to customers are referred to a named bank officer through the bank’s maker-checker process. Inference takes place in India within the bank’s environment, while data is not pooled, used to train shared models or retained by model providers. Recommendations and approvals are recorded through an immutable audit trail.
“Banks do not have an intelligence problem. They have a latency problem. Their underlying systems were built for end-of-day processing, and adding AI on top does not change when the data arrives,” said Priyanka Kanwar, Co-founder and CEO of Falcon. “We rebuilt the core so the data arrives as the transaction happens. MerlinFX puts agents to work as transactions happen, inside the system of record, while ensuring that a bank officer approves every consequential action. This is how every financial institution will operate in 2040. On Falcon it is already how they operate.”
MerlinFX supports credit cards, credit lines, loans, current and NRI accounts and wallets across banks, NBFCs, prepaid issuers, payment aggregators and fintechs.
Send news announcements/press releases to:
editor@thefoundermedia.com
