This talk provides an in-depth look at the core principles for developing effective and dependable Generative AI (GenAI) agents. We will begin by exploring fundamental LLM Ops best practices, such as caching and latency-aware design. We will then delve into techniques for optimizing agent performance, including reranking, session management, and advanced prompt engineering. Additionally, the session will cover critical aspects of GenAI development like Reinforcement Learning from Human Feedback (RLHF), Supervised Fine-tuning (SFT), and robust evaluation methods. Attendees will leave with a comprehensive understanding of the architectural considerations essential for constructing high-performing and reliable GenAI agents.

This workshop examines the critical lessons learned in deploying generative AI applications and explores how these insights can shape the future of the "AI Agents era." We'll delve into the practical challenges and solutions encountered in moving from initial answers to actionable implementations, prescribing how these learnings can be applied to build more sophisticated, autonomous, and impactful AI systems.

The pond became a crime scene, and Gemini was the detective! Join us as we share how Gemini was used to investigate the mysterious deaths of some fish. Discover how this powerful large language model (LLM) can analyze hours of video footage to identify threats, automate responses, and help solve real-world problems – all without any pretraining or complex setup. In minutes, we’ll show you how to analyze videos and get insightful results with just a few lines of code.

Business intelligence professionals and data practitioners, join us to share experiences and brainstorm solutions to your biggest data challenges. Whether you're tackling AI hallucinations, accelerating large queries, or defining data governance, this meetup provides a space for interactive discussion with peers and data experts. Bring your own topics to explore real-world scenarios and grow your network within our community.

Hear how generative media is enabling organizations to uplevel their content workflows and provide new and exciting features at scale. With Imagen and Veo integration, users of all skill levels are producing high quality visuals in a fraction of the time. Learn from companies L’Oreal and Honor on how their enterprises have provided access to AI-powered creative tools to their employees and customers and have delivered real business impact as a result.

Agencies are applying cutting-edge technology to safeguard communities and resources, bolster emergency management, and streamline disaster response efforts. Learn how AI, predictive capabilities, real-time data analysis, and a modernized IT infrastructure can be used to build a more resilient future.

This meetup is your forum to connect with fellow Workspace admins, share challenges, and discover solutions. Engage in peer-to-peer learning, troubleshoot common issues, and stay ahead with the latest Workspace updates. Come ready to contribute, learn, and build a stronger Workspace admin community!

Overwhelmed by the complexities of building a robust and scalable data pipeline for algo trading with AlloyDB? This session provides the Google Cloud services, tools, recommendations, and best practices you need to succeed. We'll explore battle-tested strategies for implementing a low-latency, high-volume trading platform using AlloyDB and Spark Streaming on Dataproc.

Leverage Composer Orchestration to create a scalable and efficient data pipeline that meets the demands of algo trading and can handle increasing data volumes and trading activity by utilizing the scalability of Google Cloud services.

Effectively managing the balance between granular access control and administrative overhead is an ongoing challenge in GCP. Cloud IAM offers granular permissions, but enforcing complex access requirements like mutual exclusivity is not straightforward. Cloud PAM enables just-in-time permission provisioning, but lacks native mutual exclusivity enforcement.

This session dives into the practical application of generative AI and Google Cloud tools for transforming computer science education. Discover how educators can integrate these technologies into core computer science courses to equip future developers with the in-demand skills needed for an AI-driven future. We'll explore the Google Tech Exchange curriculum, a model for bridging the gap between academia and industry through hands-on experience with cutting-edge AI and cloud technologies.

Multimodal AI is transforming how users interact with applications. Join us for an exploration of how to build secure, scalable, and intelligent multiplatform and multimodal AI apps using Flutter, Firebase, and Gemini. Leveraging Gemini's advanced multimodal understanding, we'll deconstruct the architecture of a cutting-edge app, demonstrating how to seamlessly integrate voice, image, and text interactions into your web and mobile experiences using a single codebase. This deep dive into architecture and code, along with Q&A, will empower you to create engaging user experiences powered by the latest advancements in AI, backed by Google Cloud.

Calling all Google Workspace developers! Join us for a lively discussion on Apps Script, the cloud-based JavaScript platform that empowers you to automate tasks, connect services, and extend the functionality of your favorite Google apps. Whether you're a seasoned Apps Script pro or just starting out, this session offers a space to share knowledge, exchange tips, and connect with fellow enthusiasts.

Distributed computing for ML is hard. JAX makes it easy (with TPUs & GPUs). Come share your experiences with JAX so we can all learn when JAX is the easy answer and when you should be looking for alternatives.

This talk explores how Google Batch accelerates genomics analysis on Google Cloud with two powerful solutions: 1. Accelerated DRAGEN with Google Batch and Eviden FPGAs: Combines Illumina DRAGEN bioinformatics pipelines with Google Batch and Eviden FPGAs for speed, scalability, and cost-efficiency in secondary analysis. 2. Scalable Genomics Workflows with NVIDIA Parabricks and Google Batch: Simplified GPU management, high-throughput analysis, and cost-effective scalability by automating GPU deployment and management with Google Batch.

This talk focuses on using conversational AI to drive revenue growth and optimize customer experience in your contact center. Discover how AI-powered virtual agents can proactively engage customers, identify sales opportunities, qualify leads, and seamlessly hand off high-priority interactions to human agents. Additionally, explore strategies for implementing tiered service models (freemium and premium) to enhance customer journeys and increase revenue. Premium service offerings could include faster response times, queue bypassing, and video calls with live agents.

Step into the journey from startup to IPO with exclusive insights from Spotify. Discover how to navigate pivotal moments on the growth path, from evolving your organizational structure to tackling major architectural shifts that prepare you for an expanding customer base. Learn how cutting-edge AI can be your ally in streamlining operations, scaling with precision, and making data-driven decisions that set you apart.

How to successfully migrate petabyte-scale of Cassandra clusters to Spanner without requiring code changes. The use case addresses various lifecycle aspects, including IaC, containerization, gradual migration, performance testing, security, centralized observability and multi-region operations. 

Learn from public sector leaders about how they are innovating and improving the citizen experience with AI-powered tools and virtual agents. By applying technology to streamline processes, personalize interactions, and improve access to information, public sector agencies are able to more efficiently and effectively serve their citizens.

Join this session where Shopify engineers will discuss how they leverage the latest Google Kubernetes Engine (GKE) innovations to build robust, scalable platforms that not only handle everyday traffic with ease but also gracefully absorb unpredictable spikes during peak events like Black Friday and Cyber Monday. Learn key architectural patterns, smart infrastructure choices, and proven best practices. Discover how to optimize resource utilization, control costs, and deliver cost-effective performance every time.

As the number of distributed and generative AI applications grows in your organization, so will the number of APIs – presenting management and governance challenges. This session introduces API Hub, a solution for unified API management. Learn how to gain centralized observability over your APIs, use the API Hub on-ramp framework to connect to any API gateway, help developers access shared components and standards to build compliant APIs faster, and discover and manage shadow APIs.

This workshop examines the critical lessons learned in deploying generative AI applications and explores how these insights can shape the future of the "AI Agents era." We'll delve into the practical challenges and solutions encountered in moving from initial answers to actionable implementations, prescribing how these learnings can be applied to build more sophisticated, autonomous, and impactful AI systems.

Learn how a financial services company successfully transitioned its petabyte-scale logging platform from an on-premises environment to a scalable, cost-effective, and secure solution on Google Cloud. This talk will delve into the key phases of their journey, including rigorous benchmarking, scale testing, and the implementation of seamless log ingestion. Explore how the solution leveraged Terraform-driven automation to efficiently manage over 250,000 resources and preserved existing content such as searches, alerts, and dashboards.

Join us for an engaging meetup with AppSheet Developer experts who'll be on hand to share their top tips and tricks in building Enterprise quality solutions.

Commerzbank, the largest bank in Germany, is using Gemini to transform its financial advisory workflows. The company is seeing significant reductions in processing time of client documentation and increases in productivity with Vertex AI. Learn how Gemini's ability to understand multiple modalities and process long documentation along with the generative AI evaluation service on Vertex AI, played a key role in streamlining operations and enhancing the client experience.

Java developers, join us to connect with your peers and explore best practices for building robust GenAI Java applications. We'll walk through sample apps as we discuss topics including cost reduction, efficient embedding generation, entity management, and testing. Then we'll dive deeper into the topics most relevant to you.