Language models have already evolved to do much more than language tasks, principally in the domain of image, audio, and soon video. Join Mostafa Dehghani to explore the emergent frontier of multimodal generation, what Gemini’s world knowledge unlocks that domain specific models cannot create, and how developers should be thinking about AI as a next-generation creative partner.

Join us for an insightful discussion with Understood.org, a leading nonprofit dedicated to supporting the 70 million Americans with learning and thinking differences, and discover how empowering neurodiverse employees with tools like Google Workspace and Gemini can foster a more productive workplace.

Learn how Database Migration Service can help you modernize your SQL Server databases to unleash the power of cloud databases and open source PostgreSQL! Convert your SQL Server schema and T-SQL code to PostgreSQL dialect with a click of a button in the DMS Conversion Workspace. Some objects could not be fully converted? Gemini can suggest a fix. Not yet familiar with PostgreSQL features? Ask Gemini to teach you how to convert SQL Server features to PostgreSQL equivalent ones. While Gemini is there - ask it to optimize the converted code or add some comments to explain the business logic. Once your database is fully converted and optimized you can migrate the data with minimal downtime using the change data capture powered migration job and complete your migration journey.

AI is revolutionizing observability. Learn about Cloud SQL AI-powered Database Insights and how it can help you optimize your queries and boost database performance. We’ll dive deep into the new Insights capabilities for MySQL, PostgreSQL, and SQL Server, including the Gemini-powered chat agent. Learn how to troubleshoot those tricky database performance issues and get practical tips to improve the performance of your applications.

Gemini was built from the ground up to support our breakthrough long context window, with up to 2 million tokens in our largest models. Join Nikolay Savinov to explore how to get the most out of long context and what a world of infinite context might look like.

Join us for an interactive session where we’ll build, deploy, and scale inference apps. Imagine creating and launching generative AI apps that deliver personalized recommendations and stunning images, all with the unparalleled efficiency and scalability of serverless computing. You’ll learn how to build gen AI apps effortlessly using Gemini Code Assist; deploy gen AI apps in minutes on Cloud Run, using Vertex AI or on-demand, scale-to-zero serverless GPUs; and optimize the performance and cost of AI workloads by implementing best practices.

Learn how to evaluate and optimize the impact of AI-assisted software development with Gemini Code Assist. This session covers processes for measuring AI-assistance effectiveness, exploring quantitative and qualitative measures available with Gemini Code Assist, and integrating with Cloud Monitoring and Cloud Logging. Discover how to leverage DevOps Research and Assessment (DORA) metrics to track productivity gains. Whether you’re a developer, team lead, architect, or IT manager, you’ll gain insights into measuring the impact of AI assistance.

AI-enabled browser agents are in the news now, but it’s not always clear how they solve real-world problems. In this session, we’ll share our experience building a web browser agent by integrating Gemini into an end-to-end service that follows text instructions to take actions in a web application. We’ll take you through our journey of creating the agent, share the research that inspired us, and show how we’ve used the system to tackle practical problems like validating user flows in the UI and semantically checking web links.

This talk demonstrates a fashion app that leverages the power of AlloyDB, Google Cloud’s fully managed PostgreSQL-compatible database, to provide users with intelligent recommendations for matching outfits. User-uploaded data of their clothes triggers a styling insight on how to pair the outfit with matching real-time fashion advice. This is enabled through an intuitive contextual search (vector search) powered by AlloyDB and Google’s ScaNN index to deliver faster vector search results, low-latency querying, and response times. While we’re at it, we’ll showcase the power of the AlloyDB columnar engine on joins required by the application to generate style recommendations. To complete the experience, we’ll engage the Vertex AI Gemini API package from Spring and LangChain4j integrations for generative recommendations and a visual representation of the personalized style. This entire application is built on a Java Spring Boot framework and deployed serverlessly on Cloud Run, ensuring scalability and cost efficiency. This talk explores how these technologies work together to create a dynamic and engaging fashion experience.

Join Woosuk Kwon, Founder of vLLM, Robert Shaw, Director of Engineering at Red Hat, and Brittany Rockwell, Product Manager for vLLM on TPU, to learn about how vLLM is helping Google Cloud customers serve state-of-the-art models with high performance and ease of use across TPUs and GPUs.

Learn how LG AI Research uses Google Cloud AI Hypercomputer to build their EXAONE family of LLMs and innovative Agentic AI experiences based the models. EXAONE 3.5, class of bilingual models that can learn and understand both Korean and English, recorded world-class performance in Korean. The collaboration between LG AI Research and Google Cloud enabled LG to significantly enhance model performance, reduce inference time, and improve resource efficiency through Google Cloud's easy-to-use scalable infrastructure

AI application growth brings security challenges, especially in multi-tenant environments. This talk explores access control issues in agentic workflows, using Google's Reasoning Engine. We'll cover limitations of current methods, user/agent identity interplay, data isolation, and secure delegation to LLMs. Future solutions include fine-grained access control and secure identity propagation. Learn to design secure, scalable AI workflows.

Experience a new way to interact with LLM-powered agents! With Gemini 2.0 and Multimodal Live API, users can give audible instructions and show visual content from a camera or screen, while receiving spoken responses from the model. This enables more natural, timely communication and unlocks multimodal agent workflows. This session showcases how existing agent experiences can be adapted for voice and visual cues, and explores new possibilities with this technology.

Building a GenAI chatbot for millions of users? This session reveals the secret sauce: best practices in LLM orchestration, advanced agentic patterns, and grounded responses, all while prioritizing brand safety. Learn key architectural decisions for balancing latency and quality, and discover strategies for scaling to production.

This hands-on lab introduces Gemini 2.0 Flash, the powerful new multimodal AI model from Google DeepMind, available through the Gemini API in Vertex AI. You'll explore its significantly improved speed, performance, and quality while learning to leverage its capabilities for tasks like text and code generation, multimodal data processing, and function calling. The lab also covers advanced features such as asynchronous methods, system instructions, controlled generation, safety settings, grounding with Google Search, and token counting.

If you register for a Learning Center lab, please ensure that you sign up for a Google Cloud Skills Boost account for both your work domain and personal email address. You will need to authenticate your account as well (be sure to check your spam folder!). This will ensure you can arrive and access your labs quickly onsite. You can follow this link to sign up!

Build a multimodal search engine with Gemini and Vertex AI. This hands-on lab demonstrates Retrieval Augmented Generation (RAG) to query documents containing text and images. Learn to extract metadata, generate embeddings, and search using text or image queries.

If you register for a Learning Center lab, please ensure that you sign up for a Google Cloud Skills Boost account for both your work domain and personal email address. You will need to authenticate your account as well (be sure to check your spam folder!). This will ensure you can arrive and access your labs quickly onsite. You can follow this link to sign up!

Discover how leading manufacturing companies are leveraging AI to transform employee experience and drive productivity with AI-enabled workflows and assistants. Learn about the evolution of these tools, the positive user feedback they received, and further possibilities around image and code generation. This session will provide a compelling look at how AI can revolutionize large enterprises, with practical insights into implementation and future possibilities.

Generative AI is transforming every role in your organization. In this panel, Google and AI platform leaders Salesforce and Atlassian will share how Gemini for Google Workspace and partner integrations can transform how your organization works. And Wayfair will share how they’re using Gemini and Salesforce to drive their business. Discover how Workspace empowers the future of work with Gemini integrated into how every role gets work done, boosting productivity and unlocking growth.

Unleash the power of AI-driven analytics with Gemini and BigQuery. Learn how to uncover trends, automate complex queries, and make smarter decisions. Explore real-world examples from industry leaders using AI to extract insights from unstructured data like text, images, and audio to streamline operations and unlock growth.

Struggling to monitor the performance and health of your large language model (LLM) deployments on Google Kubernetes Engine (GKE)? This session unveils how the Google Cloud Observability suite provides a comprehensive solution for monitoring leading AI model servers like Ray, NVIDIA Triton, vLLM, TGI, and others. Learn how our one-click setup automatically configures dashboards, alerts, and critical metrics – including GPU and TPU utilization, latency, throughput, and error analysis – to enable faster troubleshooting and optimized performance. Discover how to gain complete visibility into your LLM infrastructure.