Hands-on guidance for securing an MCP server on Google Cloud.

Découvrez Gemini CLI, un outil en ligne de commande permettant d'interroger et de modifier des bases de code de toute taille. Cette démonstration en direct présente des workflows pratiques pour l'intégration de code existant, le développement de fonctionnalités, la CI/CD, ainsi que l'automatisation de vos tâches d'ingénierie quotidiennes.

Abdel est un expert en développement cloud chez Google Cloud. Il co-anime le podcast Kubernetes de Google et est ambassadeur de la CNCF. Ses domaines d'expertise sont GKE/Kubernetes, les Service Mesh et le Serverless. Abdel a débuté sa carrière dans les centres de données et l'infrastructure au Maroc, son pays d'origine, avant de rejoindre le plus grand centre de données européen de Google en Belgique. Il a ensuite intégré l'équipe Google Cloud Professional Services en Suède et a travaillé pendant cinq ans avec les clients de Google Cloud sur l'architecture et la conception de systèmes distribués à grande échelle, avant de se consacrer à la promotion et à l'engagement communautaire.

Dans cette présentation, nous explorerons le protocole MCP (Model Context Protocol) et sa mission pour standardiser la communication entre nos applications et les LLMs. Nous commencerons par examiner les composants essentiels à la communication entre un client et un serveur, les différents protocoles de transport et les éléments constitutifs, notamment les outils, les prompts et les ressources. Nous aborderons ensuite des cas d'utilisation concrets, illustrant l'impact déjà significatif de MCP pour coder et intégrer de l'IA. Enfin, nous verrons comment l'écosystème Go adopte ce protocole pour permettre un développement MCP robuste et une nouvelle génération d'expériences de programmation.

Composing content for the first time has always been a synchronous, uninterruptible process. The performance of this model is not ideal for computationally expensive compositions, affecting lazy layouts and heavy screen transitions the most. In the next release of Compose runtime we are introducing pausable composition that addresses this problem. It enables finer control over initial composition, allowing for much more efficient scheduling. This talk provides a deep technical overview of the feature, discussing implementation details and capabilities it unlocks. You will learn why pausable composition is the hottest new feature in Jetpack Compose and how your apps can benefit from it.

Workshop: TCS + Partners

Talk on rapid evolution of agent-to-agent systems.

Kick-off and briefing for the AI Challenge.

Executive roundtable discussion on how platforms act as the foundation layer for AI initiatives. Topics include the implications of the latest DORA report and State of AI in Platform Engineering, the AI–platform engineering relationship, and lessons from top-performing organizations on unlocking real value from AI initiatives using platforms.

This session dives into an AI-accelerated development technique for building and deploying a Gmail Add-on. We'll explore integrating Gemini CLI for AI-assisted development, Google Apps Script for backend logic and Workspace integration, and the Model Context Protocol (MCP) to connect Gemini CLI to Google Workspace tools and Google Workspace developer documentation. Join us to gain the practical knowledge needed to extend Gmail's capabilities with modern generative AI features.

The talks introduce GEE with a focus on the latest remote sensing foundational model from GDM, AlphaEarth Foundations. Unified Data Representation: They integrate vast and diverse Earth observation data—including optical satellite images, radar, 3D laser mapping, and climate simulations—into a single, unified embedding that computers can easily process. High Efficiency and Detail: The embeddings are highly compact, requiring 16 times less storage space than other tested AI systems, which dramatically reduces the cost of planetary-scale analysis while still characterizing the Earth in sharp 10x10 meter squares. Accurate, On-Demand Mapping: This technology enables scientists to create detailed and consistent maps on-demand with high accuracy, even in areas with sparse data, to better monitor critical issues like deforestation, urban expansion, and water resources.

This session provides a comprehensive look at automating and integrating with Google Workspace. First, we'll cover the latest AI and Workspace Flows capabilities that enhance efficiency and collaboration across your organization. Then, we'll shift focus to the powerful developer opportunity within Workspace Flows. You will gain a deep understanding of the frameworks for extending Flows, with a practical walk-through on how to build a custom step using the Google Workspace Add-ons framework. Attend to hear specific use cases and leave with the knowledge needed to create powerful, tailored automation and integration solutions.

Hands-on workshop led by Google Workspace Developer Relations team.