Tracing AI Calls End to End: What AI inference teaches us about... Jigar Kuverji Savla, Tensordyne

Why watch

This talk explores how gRPC service boundaries can serve as the backbone for end-to-end AI inference tracing. It covers which telemetry signals matter across AI-serving workflows, how to distinguish prefill-heavy from decode-heavy behavior, and how to expose telemetry that helps ML and infra teams understand what happened during inference requests. Designed for platform engineers and ML infrastructure teams working on production AI systems.