Hemalatha Murugesan is a Software Engineer and Independent Researcher with over 12 years of experience in enterprise software development, specializing in core banking systems, commercial lending platforms, contact center technologies, and cloud-native applications. She has led and contributed to the design and implementation of large-scale banking solutions, including commercial lending, collections, IVR modernization, and customer engagement platforms using technologies such as Java microservices, cloud computing, and AI-driven automation. Her research interests include intelligent software architecture, artificial intelligence, machine learning, fraud detection, predictive analytics, and digital transformation in the banking and financial services sector. She is an IEEE member, an active peer reviewer for international conferences and journals, and is committed to bridging industry experience with academic research to develop innovative, scalable, and secure financial technology solutions. Hemalatha is passionate about advancing AI-powered banking systems that improve operational efficiency, enhance customer experience, and strengthen risk management through intelligent software engineering.  

Title: Intelligent Software Architecture for Real-Time Fraud Detection in Commercial Lending Using AI and Cloud-Native Microservices -  Abstract:  The rapid evolution of digital banking has significantly increased the complexity of fraud detection in commercial lending systems. Traditional rule-based fraud detection solutions often struggle to identify sophisticated fraud patterns in real time while maintaining the scalability required by modern financial institutions. This presentation proposes an intelligent software architecture that integrates artificial intelligence, machine learning, cloud-native microservices, and event-driven processing to enhance fraud detection across commercial lending platforms. The proposed architecture combines data from core banking applications, loan origination systems, customer interaction channels, and contact center platforms to build a unified fraud intelligence framework. Machine learning models continuously analyze transactional behavior, customer interactions, lending patterns, and historical risk indicators to detect anomalies with greater accuracy while reducing false positives. The presentation also discusses how cloud-native technologies—including containerized microservices, API-driven integration, real-time streaming pipelines, and automated deployment—improve scalability, resilience, and operational efficiency in banking environments. Real-world implementation considerations, security architecture, compliance requirements, and AI governance are also explored. This session demonstrates how intelligent software architecture can transform commercial lending operations by enabling proactive fraud prevention, improving customer trust, and supporting secure digital banking transformation.