A common criticism of agentic AI conversations is that they often feel too abstract. To address this, a researcher has shared a breakdown of their own agentic system, which aims to provide a more tangible understanding of how these systems can be applied in real-world scenarios.

The system, designed to identify and analyze cases of AI adoption within companies, utilizes six agents to find and evaluate use cases, extract key details, add context, and match them to users’ interests. These agents also maintain research logs, reporting back when they encounter obstacles.

Notably, the system relies on a shared database of cases, research logs, and human decisions, rather than relying on complex orchestration tools. This approach is seen as a potential starting point for many useful agentic systems, which can augment human judgment and facilitate scaling.

The same setup could be applied to various domains, including competitor research, real estate, supply chain management, and more. By sharing this practical example, the researcher invites thoughts and potential improvements, highlighting the importance of collaboration in advancing the field of agentic AI.

Photo by Mariano Di Luch on Pexels
Photos provided by Pexels