OpenAI Agent SDK Integration with Azure Durable Functions

Having shifted my focus from implementation to researching the practical applications of AI at scale, Im keen to understand the boundaries of this architecture.... The article clearly shows how it manages state within an orchestration. My research is currently focused on a key challenge like "state and context management across multiple, heterogeneous agents like getting multiple specialized agents (like a researcher, a negotiator and a checker) to work together effectively. How can we extend this beyond sequential tasks to handle complex teamwork and shared knowledge between agents???? This feels like a foundational piece of that puzzle🤔🤔 Thank you for sharing the detailed post✌️