Amit kumar Meshram

Big Data Engineering and Reliable Intelligence

Abstract: 

As Big Data and AI systems increasingly power customer-facing and mission-critical platforms, trust has become a fundamental engineering challenge. While many AI solutions perform well in controlled environments, they often fail under real-world conditions—where data is late, incomplete, duplicated, or out of order, and where systems must operate under strict latency and availability constraints.
This keynote examines how trust can be deliberately engineered into large-scale Big Data and AI systems through architectural and resiliency-first design. Drawing from production systems operating at internet scale, the talk explores how patterns such as Command Query Responsibility Segregation (CQRS), failure isolation, and observability-driven feedback loops enable consistent, reliable, and explainable AI behavior beyond the “happy path.”

The session highlights why many AI failures are system-level failures rather than model errors, and presents practical design principles for building intelligent systems that remain dependable under real-world complexity.

Profile:

Amit Meshram is an Executive Director and Principal Software Engineer with over 20 years of experience building and governing large-scale, mission-critical systems at enterprise scale. He specializes in big data platforms, AI-driven systems, cloud and distributed architectures, and resiliency engineering that power hundreds of millions of transactions daily. A recognized technology leader and inventor with multiple patents, Amit bridges deep engineering rigor with real-world business impact, helping organizations design systems that are not just scalable—but trustworthy, resilient, and future-ready.