Abstract: This hands-on workshop teaches the design and implementation of collaborative multi-agent systems. We'll address challenges like task decomposition, agent coordination, state management, and safety. Attendees will progressively build a Local Personal Assistant, learning transferable orchestration patterns (e.g., coordinator/specialist). We'll use Python and the open-source Agent Development Kit (ADK) to illustrate these principles, emphasizing concepts applicable beyond any single tool.
Learning Objectives:
Design multi-agent systems by decomposing complex problems.
Implement core agent capabilities (instructions, tools, basic flow control) using ADK.
Apply fundamental orchestration patterns for agent collaboration.
Manage conversational state for context sharing.
Incorporate basic safety guardrails.
Attendee Requirements:
Laptop Required: This is a hands-on coding workshop.
Software Prerequisites:
Python 3.9+ installed.
pip (Python package installer).
Ability to use a terminal/command prompt.
Bio: Ivan Nardini is a Developer Relation Engineer on Google’s Cloud team, focusing on Artificial Intelligence and Machine Learning. He enables developers to build innovative AI and ML applications using their preferred libraries, models, and tools on Vertex AI, through code samples, online content, and events. Ivan has a master degree in Economics and Social Sciences from Università Bocconi and he attended a specialized training in Data Science from Barcelona Graduate School of Economics.



