The core of this experiment is a collaborative "do-and-learn" loop where the agent executes a proven, high-stakes lead generation methodology, and the human provides the "judgment" layer.
This is an ambitious experiment exploring a few points:
Judgment vs. Execution: Humans learn good judgment through direct experience, failures, and working through problems. If we offload the doing to an agent, how do we build human-level judgment in the loop? Recursive Skill Building: Moving beyond simple Chain-of-Thought or generic skills. Can an agent document, refine, and store their own processes as they execute based on feedback? Training via Methodology: Using a proven, multi-million dollar revenue lead-generation methodology as the training data to see if an agent can move from instruction-follower to process-owner.
I’d love feedback on the architecture of the agent loops and how others are handling the "education" of their agents in complex, multi-day task or research cycles.