Vasanth

This session explores how large-scale AI systems can power end-to-end content pipelines, from intelligent generation and dynamic scheduling to automated quality assurance, across millions of customer touchpoints. Drawing from production experience at Amazon, the talk covers the architectural patterns, multi-model orchestration strategies, and evaluation frameworks that enable reliable and scalable content operations at a global scale. Attendees will gain practical insights into building robust ML platforms that balance speed, quality, and operational efficiency in real-world deployment environments.

Biography:

Vasanth Rajendran is an engineering manager at Amazon focused on building large-scale AI systems for retail discovery and personalization. He leads cross-functional engineering and science teams that design and operate automated platforms for AI-driven content generation, scheduling, and continuous quality assurance, powering millions of customer journeys globally. His work spans distributed systems, production ML platforms, multi-model content generation pipelines, large-scale scheduling systems, and automated evaluation and QA frameworks. Vasanth is an active IEEE author and frequent speaker on production AI system design, with a strong focus on reliability, scalability, and real-world deployment.

Email:

Vasanth

Biography:

Vasanth Rajendran is an engineering manager at Amazon focused on building large-scale AI systems for retail discovery and personalization. He leads cross-functional engineering and science teams that design and operate automated platforms for AI-driven content generation, scheduling, and continuous quality assurance, powering millions of customer journeys globally. His work spans distributed systems, production ML platforms, multi-model content generation pipelines, large-scale scheduling systems, and automated evaluation and QA frameworks. Vasanth is an active IEEE author and frequent speaker on production AI system design, with a strong focus on reliability, scalability, and real-world deployment.