This talk explores a solution for the analysis of medical images, with the goal of identifying subtle indicators of cancer. The solution encompasses the ingestion, analysis, annotation, and storage of medical images to facilitate the detection of cancerous cells and enable further insights. Additionally, the solution supports the development of AI diagnostic models by converting images and metadata into formats conducive to machine learning, such as embeddings.

Our customer had been using Spanner for relational data and Tiger Graph for graph data, which prevented them from cross-referencing the two and created additional processing overhead. To address this, Google migrated the customer’s graph datasets into Spanner GraphDB. This involved creating TDD, scripts, and tools for graph schema transformations and loading data into the graph DB. Google also developed the long-awaited Corporate graph hierarchy for the customer and assisted with multiple graph query use cases essential to their business.

Discover how Vertex Search Commerce revolutionizes e-commerce by delivering world-class, personalized experiences that truly delight customers. Learn how strategic migration to this intelligent search engine, focusing on robust data pipelines and embracing AI-powered ranking, can significantly reduce zero-result searches and dynamically showcase trending items. This talk delves into the critical role of A/B testing in objectively measuring the transformative power of Vertex Search. We'll explore how rigorous experimentation allows us to not only measure the positive impact but also to identify and rectify any underlying configuration issues. Discover how a data-driven approach through meticulous A/B testing ensures Vertex Search is at peak performance, enabling you to fine-tune your implementation and be at the forefront of AI-powered product discovery landscape.

Join Aritzia, Anomalo, and Google Cloud to learn how Aritzia automates data quality across 500+ sources in BigQuery. Discover how integrating Anomalo with Google Cloud helps proactively detect anomalies, maintain data integrity, and build trust in analytics. Explore how automation reduces time spent troubleshooting and increases time spent creating business value through reliable, AI-enhanced analytics.

Connect with fellow developers building and running applications in production on Cloud Run. We'll discuss the latest announcements, new features, and best practices for scaling and managing deployments. This is a great opportunity to share your real-world use cases, discuss challenges you’ve faced, and connect with others who rely on Cloud Run.

Maximize your Gen AI inference performance on GKE. This session dives into the latest Kubernetes and GKE advancements, revealing how to achieve significant cost savings, reduced latency, and increased throughput. Discover new inference features on GKE for optimizing load balancing, scaling, accelerator selection, and overall usability. Plus, hear directly from Snap Inc. about their journey re-architecting their inference platform for the demands of Gen AI.

Enterprise Architects are essential for successful cloud innovation. This session explores the evolving role of Enterprise Architecture and how to leverage Google Cloud.

Key Takeaways:
- Understand the changing role of Enterprise Architecture in a cloud-centric environment
- Learn strategic frameworks for aligning cloud adoption with business objectives
- Explore key considerations for hybrid and multi-cloud strategies
- Gain insights from real-world case studies of large-scale Google Cloud migrations

Gemini 2.0 represents a significant leap forward in image understanding. Its object detection capabilities are dramatically faster than anything before, enabling near-instantaneous identification of visual elements. Combined with Gemini's advanced reasoning and access to external tools, this speed unlocks a vast range of new applications and possibilities, from rapid image search to complex visual problem-solving. Critically, Gemini 2.0 also possesses an experimental capacity for 3D scene understanding, allowing it to interpret spatial relationships and depth unlocking a wealth of new possibilities across diverse domains.

This talk will discuss how to efficiently set up and deploy Google Cloud TPUs for training AI and LLM workloads. Using scripts and existing tools to automate the process, you'll learn how to quickly build a fully functional TPU training environment. Additionally, the session offers insights on optimizing your cloud environment for maximum training efficiency, accelerating your AI and LLM initiatives.

This meetup is for app publishers and engineers building with Flutter -- or curious about what it's like! Bring questions for the team and our community, share an app that you're proud of, and see how Flutter can power experiences that users love on the platforms they're already using.

Cloud, AI, and data analytics technologies have the potential to revolutionize healthcare - streamlining eligibility determination, automating administrative processes, and providing greater access to critical services. By tailoring interventions, resources, and communication strategies to individual needs, technology can enable more effective, engaging, and personalized care for all.

Unlocking insights often requires data collaboration between different organizations, which raises important questions about privacy and data security. Learn how Google and one of our customers use confidential computing solutions to apply new approaches to data security, isolation, and transparency – all through the use of Confidential Computing and Trusted Execution Environments (TEEs). TEEs can transform privacy obligations into real-world, technical solutions that enable businesses of all sizes to thrive while respecting people’s privacy.

This session offers a technical deep dive of the state-of-art AlloyDB AI capabilities for building highly accurate and relevant generative AI applications using real-time data. We’ll cover vector search using Google Research’s ScaNN index technology and cover how you can utilize Gemini from AlloyDB operators to seamlessly integrate into your application. Discover AlloyDB AI natural language feature, a new way to interact with databases and how it accurately and securely answers your questions. Also learn about the latest research between Google and NVIDIA on GPU-accelerated vector index builds in databases.

Businesses and public sector organizations are transitioning away from legacy digital workplace solutions to enhance productivity and collaboration, optimize IT costs and workload, and better mitigate security and compliance risks with modern, secure-by-design architecture and controls. Join this session to discover how organizations of all sizes have transformed their organization’s collaboration and improved the security posture with Google Workspace and leading identity and device management solutions from Okta, Omnissa, and JumpCloud.

"Talk-to-Manuals" leverages Google Cloud's Vertex AI Search and Conversation, along with a multi-turn GenAI agent, to provide customers with easy access to vehicle information. The solution integrates and indexes diverse data sources—including unstructured PDFs and structured data from FAQs—for efficient retrieval. Semantic search, powered by LLMs, understands user intent, delivering highly relevant results beyond keyword matching. The interactive multi-turn conversation guides users towards precise information, refining queries and increasing accuracy. Data filtering based on vehicle attributes (model, year, engine type) combined with sophisticated GenAI-powered post-processing ensures tailored results. The successful implementation of "Talk-to-Manuals" significantly improved customer service by providing a more accessible and intuitive way for customers to find information.

Unlock the true potential of your e-commerce platform by delivering exceptional product discovery experiences. This session delves into the nuances of transitioning to intelligent search with Vertex AI Search for Commerce. Discover how to architect robust and scalable data pipelines for catalog and event ingestion, the very foundation for personalized product discovery and delighted users. We'll explore the anatomy of critical data like search events and demonstrate how a thoughtful approach to data ingestion, which directly fuels the quality of your search results. More than just search, we'll unveil how combining these APIs unlocks opportunities for implementing creative and engaging features, paving the way for innovative experiences like conversational commerce. Join us to learn how to transform your search into a revenue-driving engine.

This talk offers a solution to accelerate healthcare innovation by streamlining the conversion and integration of various data formats (HL7 v2, CSV, RDBMS, etc.) into the FHIR standard.

This solution reduces the need for manual mapping allowing for quick conversion of various healthcare data formats into FHIR and significantly reduces the workload of healthcare IT teams. FHIR data is then loaded into Google BigQuery providing a scalable and secure platform for data storage and analysis.

Join GXO's SVP of Data/AI/ML, Ramin Rastin, and Senior Director of Data, Americo, as they share their remarkable journey of implementing Generative AI and Vertex AI within their organization. Discover how GXO leveraged Google Cloud's cutting-edge technology to achieve significant scalability and efficiency gains, enhance business operations, and improve user experience. In this session, you'll learn how GXO: Scaled rapidly: Deployed 15-20 AI-powered assistants within the first six months of 2025, each reducing workloads by 40-60%. Enhanced flexibility: Developed a Generic Chat Assistant mid-project to adapt to evolving business needs. Improved accessibility: Created an intuitive user interface for all users, regardless of technical expertise. Centralized control: Implemented an Admin Page with role-based access control for enhanced security and governance.

AI applications that use the retrieval augmented generation (RAG) architecture can be a lot more accurate than those built with just the AI model’s built-in knowledge. But RAG introduces additional processing steps which need to be monitored. We’ll show you how.

This meetup is a space for developers actively working with any open-source AI libraries, frameworks, or tools, to share their projects, challenges, and solutions. Whether you're building with LangChain, Haystack, Transformers, TensorFlow, PyTorch, or any other open-source AI tool, we want to hear from you. This meetup will provide an opportunity to connect with other developers, share practical tips, and get inspired to build even more with open-source AI on Google Cloud. Come ready to contribute, and let's learn from each other!

In today’s complex landscape, APIs can reside anywhere – on premises, in the cloud, or across multiple cloud providers. Join us to discover how API Hub, powered by its innovative on-ramp framework and app integration, delivers a truly unified view of your entire API ecosystem. This session is for practitioners looking to learn how to better discover, manage, and secure all their APIs, regardless of location, with comprehensive analytics and consistent governance policies.

Learn how "Project SEALD" – a collaboration between Google and AI Singapore – is building LLMs for the region. Discover why cultural context matters and how you can implement similar solutions.

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.

Join an insightful fireside chat with Jeff Dean, a pioneering force behind Google’s AI leadership. As Google's Chief Scientist at DeepMind & Research, Jeff will share his vision on AI and specialized AI hardware, including Cloud TPUs seventh generation chip; Ironwood. What exciting things might we expect this to power? What drives Google’s innovation in specialized AI hardware? In this spotlight, we’ll also discuss how TPUs enable efficient large-scale training and optimal inference workloads including exclusive, never-before-revealed details of Ironwood, differentiated chip designs, data center infrastructure, and software stack co-designs that makes Google Cloud TPUs the most compelling choice for AI workloads.

Explore how startups are creating impact for a wider range of users. Discover real-world examples of AI transforming hiring practices, product design, and user interfaces, and join the discussion on creating AI that serves everyone.