I will emphasize motivation rather than the technical details, which are (deliberately) not particularly original.
We explain why Kodamai believes Applied Category Theory is very important for the future of AI. There is so much activity in AI at the moment but much of it is driven by the low level architectural models. For example, the concept of workflows is dominant in Agentic AI but was deemed too low level to form a general model of computation by the programming languages community in the 1980s! We argue in contrast for a structure-driven approach and, consequently, a category-theretic approach.
From distributed systems and business processes to cryptographic protocols, categories emerge as a tool to refactor the models used to formalize them. In this talk, we explore how I keep finding categories in the wild, and how to use them to build guarantees.
Website based on a bootstrap design by Hartmut Eilers and Eric Koskinen.