Google I/O Extended George Town 2025

Name: Google I/O Extended George Town 2025
Start: 2025-07-26T12:30:00+08:00
End: 2025-07-26T19:00:00+08:00
Location: Masco Co-working Space

Masco Co-working Space, Jalan Dr Lim Chwee Leong, George Town, 10100

GDG George Town

Google I/O Extended is a community-led series of events that build upon Google's annual developer conference, I/O, by bringing the excitement and learning to local communities.

Jul 26, 4:30 – 11:00 AM (UTC)

0 RSVP'd

Key Themes

AIFirebaseGeminiGemmaGoogle I/O Extended

About this event

Join us for Google I/O Extended George Town 2025, a dynamic afternoon dedicated to exploring the latest advancements in artificial intelligence and cloud technology. Connect with 300 fellow tech enthusiasts and learn from a diverse lineup of experts from across Southeast Asia. This year's sessions will dive deep into the capabilities of Google's most powerful models and platforms, including hands-on insights into MedGemma for medical AI, building intelligent applications with the Gemini API and Firebase, and the future of autonomous, agentic AI. Whether you're a seasoned developer or just starting your tech journey, I/O Extended George Town offers a unique opportunity to learn, network, and get inspired.

When

Saturday, July 26, 2025
4:30 AM – 11:00 AM (UTC)

Where

Masco Co-working Space
Jalan Dr Lim Chwee Leong
George Town, 10100

Agenda

4:30 AM	Registration
5:00 AM	Opening
5:20 AM	[Hall 1] (1 hour - Talk) Agentic AI with Gemini 2.5 Pro by Kuan Hoong Poo	As AI systems grow more capable, the shift from reactive tools to agentic, goal-directed systems is redefining how we build and interact with intelligent technologies. In this session, we’ll explore the frontier of agentic AI using Gemini 2.5 Pro, Google's latest large language model. You’ll learn how Gemini 2.5 Pro enables autonomous decision-making, dynamic planning, tool integration, and contextual reasoning across complex tasks.
5:20 AM	[Hall 2] (40 mins - Talk) Securing AI Agents by Gregory Tan	As AI agents become more autonomous and integrated into real-world systems, securing them is critical. This talk explores the latest threats facing AI agents and how to defend against them. You'll learn practical strategies for securing agent workflows using Google’s Agent Development Kit (ADK) to reduce risk and improve trust. Whether you're building with Gemini or any LLM, you'll gain insights into protecting agents at every layer
6:00 AM	[Hall 2] (2 hour - Workshop) Building a Social Event Platform with ADK, A2A, MCP and Spanner Graph by Christine Tee	This workshop guides you through the Google ADK, from foundational agent development and custom tool integration (Model Context Protocol) to designing collaborative multi-agent systems. Discover how to leverage Agent-to-Agent communication and confidently deploy your agents on Google Cloud Platform.
6:20 AM	[Hall 1] (1 hour - Talk) MedGemma: Open Model for Medical AI by Lai Fong Leong	MedGemma is Google’s most advanced open multimodal model for advancing medical AI capabilities. In this session, you will discover MedGemma model overview and common use cases. Next, you will learn how to get started using MedGemma model.
7:20 AM	[Hall 1] (40 mins) Break
8:00 AM	[Hall 1] (1 hour - Talk) From Code to Conversation: Building and Deploying Multimodal AI Agents by Esther Irawati Setiawan	The next generation of AI applications will not just process text; they will see, hear, and interact with the world in real-time. This session is a comprehensive guide for developers looking to build and deploy these sophisticated multimodal experiences. We'll start with the foundation, showcasing how Google Cloud simplifies development with updates for easier deployment and using the new autonomous AI coding agent that accelerates your workflow. Then, we'll dive into the cutting edge with the Agent Development Kit (ADK), your toolkit for creating intelligent agents. The session will culminate in exploring how to harness the power of multimodality by leveraging the real-time, multimodal capabilities. You'll learn how to integrate its API, now part of Gemini Live, to build intuitive and interactive applications that can understand and respond to a dynamic world, creating truly conversational and context-aware user experiences.
8:00 AM	[Hall 2] (1 hour - Workshop) Building AI-Powered Applications with Gemini AI by Adkham Zokhirov	Gemini AI is Google's most advanced multimodal AI model, capable of understanding and generating text, images, and code. In this session, we will explore how developers can integrate Gemini AI into their applications using the Gemini API. We will cover prompt engineering, use cases such as chatbots and content generation, and best practices for optimizing Gemini-powered solutions.
9:00 AM	[Hall 1] (1 hour - Talk) MCP and A2A - what does that mean for us developers? by Weiyuan Liu	Running through a proposed project (still being constructed), using multiple agent integrations with MCP on common tools, and then using A2A to connect with multiple agents as a central orchestrator. The talk should illustrate how these two protocols work, and why they need larger companies to help them reach critical mass.
9:00 AM	[Hall 2] (1 hour - Workshop) Building AI Powered Apps with Firebase Studio with Gemini API and Genkit by Shang Yi	Learn how to use Firebase Studio to prototype AI powered applications. This workshop will also cover how Firebase Genkit supercharges AI development.
10:05 AM	[Hall 1] (45 mins - Panel Discussion) TBC
10:50 AM	[Hall 1] Closing and Lucky Draw