Point! Just thinking, agents are coming more and more to light. In example of deepseeks kind and wise to reason. What I see there reasembles two different agen

Point! Just thinking, agents are coming more and more to light. In example of deepseeks kind and wise to reason. What I see there reasembles two different agents talking to each other. While reasoning of claude ai doesn't look for me like two different agents but rather one talking to himself. So, if one would set up an organization structured agent hierarchy of boss (adviser) and managers (controller), workers (thinkers), it would be possible to do human-like decision-making.