Season 2 · Episode 1472
Stop Flying Your AI Agents Blind
Move past basic token counting. Learn how to monitor AI reasoning, prevent $47k loops, and build trust in autonomous agents.
My Weird Prompts · Daniel Rosehill
March 23, 202620m 6s
Audio is streamed directly from the publisher (dts.podtrac.com) as published in their RSS feed. Play Podcasts does not host this file. Rights-holders can request removal through the copyright & takedown page.
Show Notes
In this episode, we explore the critical shift from simple LLM monitoring to the complex world of agentic observability. As AI moves from basic chatbots to autonomous agents capable of multi-step reasoning and real-world actions, the stakes have shifted from simple helpfulness to financial and operational security. We dive into the latest tools—from OpenTelemetry-native frameworks to deterministic DAG metrics—that are helping engineers monitor the "thought" process and "action layer" of AI to prevent runaway loops and data leaks.