SYNOPSIS   

While artificial intelligence continues to dominate global headlines, moving from generic AI chat models to reliable, enterprise-grade operational systems remains a major hurdle for traditional industries. Most enterprise failures stem not from the AI models themselves, but from integration friction: connecting unstructured operational data safely into legacy databases, maintaining strict data privacy, and ensuring mathematical accuracy. This session breaks down the practical engineering framework behind deploying AI in complex, high-stakes environments. Moving past the marketing noise, the talk explores how modern data ingestion pipelines, local (on-premise) AI deployment, and automated validation layers can eliminate operational bottlenecks, reduce human error, and drive measurable ROI without compromising security or regulatory compliance.


BIODATA OF SPEAKER

Mr. Koay Xian Jing

Xian Jing is an engineer, technologist, and the founder of QuarkX Sdn. Bhd., an AI systems engineering and consulting firm based in Kuala Lumpur. With an academic background from the University of Cambridge, where his research focused on computer vision frameworks for ecological remote sensing, and extensive industry experience across enterprise software (Huawei), data engineering (Axiata Digital), and fintech automation (Ant International), he specialises in bridging complex operational workflows with modern artificial intelligence architectures. His work focuses on designing domain-specific data pipelines, privacy-first local model deployments, and real-time optimisation engines for enterprise, commercial, and industrial applications.