Building Reusable & Extendable AI Agents

About this event

Learn about building and testing agents using the Agent Development Kit (ADK) by leveraging its CLI and web tool. We'll start from the absolute basics, learning how to build a simple agent and test it instantly.

From there, we will move on to extend our agents by building custom Model Context Protocol (MCP) tools. Throughout this session, we will focus on staying away from hardcoded prompts. Instead, you'll learn how to leverage clean software design patterns to build truly reusable, extendable agents that can adapt dynamically with the use of configurable prompts and plug-and-play MCP tools, shifting your development loop from a one-off script or simple notebook into a scalable, production library.

The Agenda

Introduction to the Google ADK & CLI Sandbox: A baseline deep dive into ADK primitives. Learn how to build a simple, functional agent, establish its identity, and spin up the native web tool via the CLI to test, chat, and trace your model's reasoning in real-time.

Config-Driven AI: Stop Hardcoding Your Prompts: Why embedding system instructions inside your code is an anti-pattern. Learn how to manage agent instructions as clean, configuration data so your business logic can extend dynamically without rewriting your codebase.

Enhancing Agents with MCP Tools: Learn how to make your agents truly operational. We will discuss how to enhance your agents' capabilities by building custom and using existent Model Context Protocol (MCP) tools, giving them the power to execute real-world tasks.

Designing a Reusable Agentic Library: See how to apply structural design patterns to bundle your capabilities into a shared, plug-and-play local package. We will map out how to structure your custom wrappers so your entire team can import and attach pre-built tools to any new agent with zero code duplication.

Why Attend?

Master the ADK Core: Understand how to instantiate agents and leverage the interactive CLI to visually trace and debug how your models think.

Kill Code Redundancy: Walk away with a design pattern that allows you to write a data capability once and reuse it across multiple future agent deployments.

Think Like an AI Architect: Move away from procedural scripting and learn how to structure AI systems using professional, maintainable software engineering practices.

Who Is This For?

Software Engineers & Architects looking for a structured, maintainable way to integrate Large Language Models into enterprise applications using established design patterns.

Data Engineers who want to build clean, automated agent pipelines that connect securely to Cloud Storage and Data Warehouses without spaghetti code.

Tech Leaders & Curious Developers who understand basic Python and want to graduate from a one-off script or simple notebook into a clear, no-hype roadmap for building scalable AI tools.