Mr. Kapil Kumar Goyal

Edge-to-Cloud AI Inference Pipelines: Architecting Resilient Intelligence at Scale

Abstract:

As artificial intelligence continues to integrate into real-time, mission-critical applications—from autonomous systems to smart cities—the need for robust and scalable inference pipelines has become paramount. Traditional centralized AI architectures often fall short in scenarios demanding low latency, high availability, and adaptive decision-making at the edge. This keynote introduces a resilient, multi-tiered framework for Edge-to-Cloud AI inference pipelines, designed to address the evolving demands of modern AI workloads.
Drawing from recent advancements in distributed computing, real-time data processing, and ML model orchestration, the talk presents a layered architecture that strategically partitions inference responsibilities across the edge, fog, and cloud. Key principles such as model elasticity, data gravity, resilience under constrained environments, and context-aware deployment are explored. Real-world implementation patterns and design trade-offs are discussed, highlighting how this architecture enables intelligent systems to function reliably even in decentralized, high-velocity environments.

This session also reflects on the role of semantic interoperability, dynamic model governance, and performance monitoring in sustaining pipeline effectiveness. By connecting foundational research with practical system design, this keynote aims to inspire the next wave of AI practitioners to reimagine intelligent pipelines not merely as tools for automation, but as resilient infrastructures for real-time intelligence.

Profile:

Kapil Kumar Goyal is a Senior IEEE Member and accomplished product management leader with over 11 years of experience spearheading AI and data platform innovation across global tech enterprises. Currently serving as Senior Product Manager – AI & Data Platforms at Zillow Group (California, USA), he leads initiatives at the nexus of machine learning, large-scale data architecture, and cloud-native systems.

Renowned for delivering high-impact products from ideation to scale, Kapil combines strategic product thinking with deep technical fluency in MLOps, ML infrastructure, real-time analytics, and BI tooling—enabling the development of responsible and explainable AI systems.

A Fellow of the Soft Computing Research Society (SCRS), Kapil actively contributes to the global research community as a keynote speaker, peer reviewer for IEEE journals, and judge at international AI conferences and hackathons. His research spans topics such as edge-to-cloud AI inference pipelines, scalable data lakes, and semantic system design—advancing the frontiers of trustworthy and production-ready AI.