Agentic AI Systems for Critical Workloads

Although there is no universally agreed definition of what an AI agent is, in practice several patterns are emerging that demonstrate how to coordinate and combine the capabilities of multiple AI services, in order to create AI-based systems that can accomplish more complex tasks. These Agentic Systems architectures can be grouped in 2 main categories: workflows, where LLMs and tools are orchestrated through predefined code paths, and agents, where LLMs dynamically direct their own processes and tool usage, maintaining control over how they execute tasks. Moreover testing these Agentic Systems architectures is one of the biggest challenges for the adoption in mission critical scenarios, due to their not completely deterministic nature. The goal of this talk is giving a theoretical overview of Agentic AI in general and these patterns in particular, discussing their differences and range of applicability and showing with practical examples how they can be easily implemented and tested with Quarkus and its LangChain4j extension.