video1.0<iframe src="https://www.loom.com/embed/4f5339b543da4333b444981a23249276" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/4f5339b543da4333b444981a23249276-acbefb91bf79e903.gif5647.5952In this video, I discuss the development of a basic agent using the OpenAI agents SDK, focusing on principles and frameworks that guide our approach. I explore the concept of context engineering, emphasizing the importance of managing state, context, and agency within our agents. I also introduce the idea of creating a simple agent framework that can handle tool calls and maintain a conversation loop. Viewers are encouraged to verify that they can run the provided code and to think about how they might implement similar agents in their own projects. Please take a moment to test the code and consider how these principles can apply to your work.