The podcast episodes discuss various aspects of agentic systems, including their advantages over single-query language models, the key components that make up an agentic system, and the challenges and future outlook of agentic applications.
For example, in episode #700, Shengran Hu discusses his research on Automated Design of Agentic Systems (ADAS), an approach that uses large language models to automatically design and optimize agentic systems. In episode #698, Harrison Chase, the co-founder and CEO of Langchain, covers the company's products for building and deploying agentic systems with large language models.
The episodes also touch on the importance of communication, few-shot prompting, and multimodal models in the development of agentic systems.