Becoming AI-Led: Lessons from a Payments Company's AI Journey, with Ahu Chhapgar

Architecting an AI-Led Payments Platform #

Ahu Chhapgar, CTO at Paysafe and former executive at PayPal, Mastercard, and Citi, joins hosts Igor Tomych and Dumitru Condrea to share a defining benchmark: 64% of Paysafe’s code shipped in July was AI-generated. His central message emphasizes that while AI drastically accelerates software synthesis, it does not bypass functional requirements, regulatory compliance, or security auditing. It simply reshapes where operational effort must be applied.

The Emerging Bottleneck: Scaled Code Verification #

As AI-driven output surged across engineering teams, the primary operational constraint shifted from authoring code to validating it. Chhapgar highlights an iterative remediation strategy—persistently identifying the migrating bottleneck, resolving it, and adapting to the next. He draws a direct parallel to the early era of public APIs, which required years of industry iteration before standardized architecture emerged.

Cultural Alignment over Technical Tooling #

Ahu Chhapgar candidly notes that the primary hurdle in enterprise AI adoption is human behavior, not software capabilities. The core objective lies in ensuring personnel feel empowered rather than displaced by automation. He attributes Paysafe’s seamless rollout to early executive sponsorship from the CEO, which pre-emptively eliminated organizational friction before broad deployment began.

Fintech Risk Profiles and Structural Explainability #

Operating financial infrastructure involves stringent PCI and PII compliance mandates that typical software platforms do not face. Chhapgar outlines a dual-track framework: low-risk internal utilities move rapidly with minimal friction, whereas core financial architecture requires native explainability designed directly into system foundations. Every single code commit, whether human-written or AI-generated, must pass through identical security gates prior to production deployment.

The Reality Curve of Agentic Commerce #

Evaluating autonomous agents buying goods on behalf of consumers, Chhapgar delivers a pragmatic assessment: while live transactional volume remains marginal today, long-term commercial potential is vast. Broad market adoption hinges on global alignment across card networks, merchants, and issuers to establish explicit liability frameworks for agent-driven chargebacks and disputes.

Why Listen #

For executive leaders driving enterprise AI initiatives, this episode pairs an unprecedented performance metric with a candid analysis of structural adaptation, operational friction, and why organizational culture dictates software success.