Biography
Kishore Subramanya Hebbar is a Senior Engineer at a Fortune 500 company operating stock exchanges, clearing houses, and one of the largest residential mortgage technology platforms in the United States. With over 17 years of experience at the intersection of enterprise software architecture and financial systems engineering, Kishore specializes in applying artificial intelligence to modernize large-scale distributed systems that underpin global financial markets.
An IEEE Senior Member, Kishore is an active researcher with 14 peer-reviewed publications across IEEE Xplore, Scopus, and Springer/Atlantis venues. His work on ML-assisted service boundary detection has been independently adopted by engineering teams across multiple companies for legacy modernization. A recognized voice in financial technology, Kishore has delivered keynote addresses and served as an invited speaker at multiple conferences. He serves on a Technical Advisory Board and acts as a peer reviewer for IEEE conferences as well as Scopus and Springer-indexed journals.
Abstract
Title: When AI Meets the Regulator: Building Explainable Financial Systems That Pass Audit
Abstract: Financial institutions are deploying AI at an unprecedented scale, but the same models that detect fraud in milliseconds and reduce compliance overhead by 40% are increasingly drawing scrutiny from regulators who demand what AI engineers often cannot provide: a clear explanation of why the system made a given decision.
The tension is real and growing. The U.S. Federal Reserve’s SR 11-7 guidance, the EU AI Act’s high-risk system requirements, and emerging SEC algorithmic accountability frameworks all converge on a single demand: that AI-driven decisions in financial systems be auditable, reproducible, and explainable to a non-technical examiner. Yet most production AI architectures in financial services were not designed with this requirement in mind.
Drawing on peer-reviewed research in explainable multimodal AI systems and regulatory compliance automation, as well as production experience designing AI-augmented financial infrastructure that processes millions of transactions daily, this keynote presents a framework for building financial systems that are both high-performing and audit-ready from the ground up.
Attendees will explore:
– Why post-hoc explainability tools consistently fail regulatory audits and what structural alternatives look like.
– How compliance-first API architecture reduces manual regulatory review burden by measurable margins without sacrificing system throughput.
– How organizations can instrument their AI pipelines with observability layers using tools such as OpenTelemetry that generate the audit trails regulators actually need.
На этом сайте можно отыскать ценную информацию.
https://relax-spb.ru/read-info/1145-kak-budto-vorota-v-inoy-mir-plyusy-i-minusy-zhizni-ryadom-s-zheleznodorozhnym-vokzalom-v-ekaterinburge.html