Grokking the Agentic Loop: From Chatbots to Autonomous Agents
Kwame Wright,
Jarone Wright
Many people interact with large language models (LLMs) through simple, linear chat interfaces. However, AI unlocks its true potential when these models are placed inside an agentic loop - a cycle of reasoning, acting, and observing. In this hands-on workshop, you will demystify AI agents by building a basic agentic loop from scratch using Python and the Gemini API. Participants will learn how to move beyond static prompts to create systems that can use tools, evaluate their own outputs, and iterate toward a goal. This workshop is designed to turn the magic of AI agents into a clear, hackable process.
Saturday 1400 Sutton Place / Workshop A
Kwame Wright is a software engineer working in the autonomous vehicle space, where he integrates AI to boost productivity and improve the customer experience. He holds a PhD in electrical engineering from the University of Southern California. Outside of his core work, Kwame is fascinated by the intersection of math, physics, and art, frequently exploring topics in 3D computer graphics like path tracing and realistic rendering.
Jarone Wright is an educational technologist and computer scientist working in higher education. He holds a master’s degree from the Interactive Telecommunications Program (ITP) at the NYU Tisch School of the Arts. Outside of his academic work, Jarone focuses on creative and tangible technology, with personal interests that include physical computing, interactive music, and kiln glass.
