Mario is a senior principal software engineer at IBM working as Drools project lead. Among his interests there are also high performance systems and generative AI, being an active contributor of widely adopted projects like Quarkus and LangChain4j. He is also a Java Champion, the JUG Milano coordinator, a frequent speaker and the co-author of "Modern Java in Action" published by Manning.
According to the traditional economic theory, markets are fully efficient and humans operate in them in a rational way. In the late 70s Daniel Kahneman and Amos Tversky started disproving this efficient markets hypothesis, contrasting the consistently logical Homo Economicus (Econ) they depicted, with the more realistic Human who takes decisions based on his questionable points of view. Doing so they gave birth to the study of the psychological factors involved in the making of these decisions, called Behavioral Economics.
The same flawed reasoning also impacts other fields like software engineering: we cannot behave as cold Econ when spending or investing our money, or as rational Engeen when coding. We are humans and this inevitably influences our choices.
The anchoring effect and the availability bias affect how we benchmark and evaluate the performances of our programs. The pro-innovation and bandwagon biases drive our technical decisions, making us to blindly follow hypes and gurus. The not-invented-here syndrome pushes us to create homemade tools instead of using de-facto standards. The framing effect makes us solving the same problem in different ways, depending on how it is presented.
During this talk we will go through these and other heuristics and shortcuts used by our brain, as found by behavioral economists in almost 50 years of research, and examine them in the context of software engineering, discussing their consequences on the quality of our work.
The last few months have seen the rapid evolution of LLMs from passive completion engines, only good for generic chatbots, to components of a more complex, programmatically defined, workflow and finally into active and autonomous elements capable of reasoning, planning, and taking actions.
But moving from basic prompt engineering to truly autonomous systems requires a new class of design patterns and possibly a framework allowing to implement those patterns and put them at work in a convenient and effortless way.
In this workshop, we will explore the architecture and implementation of agentic AI using LangChain4j, a Java-native framework for building LLM-powered applications. You’ll learn how to move beyond the plain usage of a standalone LLM to design intelligent, modular agents capable of dynamic decision-making, memory retention, tool usage, RAG, MCP and A2A integration and multi-step goal execution.
After having covered the core concepts of agentic AI, we will guide you in incrementally building and testing an agentic system from scratch using LangChain4j and Quarkus, backed by real-world examples and live coding. Whether you're exploring agentic AI for task automation, intelligent assistants, or decision-support systems, this session will give you the practical tools and architectural understanding to build robust and maintainable autonomous agents in Java.
Searching for speaker images...