Setting the Stage for Agentic AI: A Practical Framework

In this episode of the AI Explored podcast, host Michael Stelzner interviews Christopher S. Penn, Chief Data Scientist for Trust Insights, about agentic AI. They discuss the misconceptions surrounding agentic AI, defining it as AI that autonomously takes actions and delivers results. Penn introduces a framework of "done by you," "done with you," and "done for you" to explain the spectrum of agentic AI implementation. The conversation covers practical tools and examples, including Google's Opal, Claude's skills, OpenAI, and N8n, highlighting their capabilities and limitations in creating AI workflows. Penn emphasizes the importance of prompt engineering and continuous testing when building AI agents, and shares a real-world example of using Claude code to write a book. The episode concludes with a discussion on the future of AI and the increasing importance of data and processes in a software-commoditized world.