ucsandman

↗ HN profile [ 55.5 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. I kept finding out what my agents did after the fact. Logs, traces, error reports. By then the action had already happened. I wanted a way to stop the action before it executes, present context to a human, and record…

  2. Hi HN, I’ve been experimenting with autonomous agents for the past year, and I kept running into the same uncomfortable problem: Agents were making decisions I couldn’t fully see or justify. They could call tools,…