AI Enabled Observability Explainers - We Actually Did Something With AI!

This keynote presentation by Vijay Samuel, Principal MTS and Architect at eBay, demonstrates practical applications of AI in cloud native observability beyond theoretical concepts. Discover how eBay's Observability platform team developed "Explainers" for telemetry signals by combining carefully crafted algorithms with Large Language Models (LLMs) to create more predictable and accurate AI-enabled experiences. Learn about their innovative approach that goes beyond simply feeding data into LLMs, including techniques for explaining traces, metrics, and logs with greater reliability. The presentation also covers how these building block explainers were combined to create compound explainers capable of interpreting entire dashboards, and explains why methods like critical path detection paired with LLMs produce superior results compared to processing raw trace data alone.