SyncAI.news, a Varaisys broadcasting
How to Find the Agent Failures Your Evals Miss with Scott Clark - #767
SC

Sam Charrington

· 53 Minutes

PodcastThe TWIML AI Podcast

How to Find the Agent Failures Your Evals Miss with Scott Clark - #767

Listen

In this episode, Scott Clark, co-founder and CEO of Distributional, joins us to explore how teams can reliably operate and improve complex LLM systems and agents in production. Scott introduces a Maslow’s hierarchy of observability: telemetry for logging, monitoring for known signals, and post-production or online analytics to surface unknown unknowns. We dig into examples of real-world failures Scott’s team has seen in production systems, such as “lazy” tool-use hallucinations that standard evals miss, and how mapping traces into vector fingerprints enables clustering and topic discovery to uncover emergent behaviors. Scott explains how analytics can feed the data flywheel by generating evals, guardrails, and training data, and why online, adaptive approaches are essential for non-stationary models. We also touch on practical how-to’s such as instrumentation with OpenTelemetry, the GenAI semantic conventions, and the role of dedicated analytics tools.

The complete show notes for this episode can be found at https://twimlai.com/go/767.

Original source

This story was published by The TWIML AI Podcast and written by Sam Charrington. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on twimlai.com

Similar News