At this year’s Google I/O conference, NVIDIA and Google Cloud are accelerating the efforts of the company’s more than 100,000 developers. Co-developers communityoffers curated learning paths, hands-on labs, and events to help you build with the full-stack NVIDIA AI platform on Google Cloud.
Announced at Google I/O last year, community A gathering of developers, data scientists, and machine learning engineers looking to sharpen their AI skills with the latest NVIDIA and Google Cloud technologies.
This year we’re rolling out new additions to the community, including a learning path for using JAX libraries on NVIDIA GPUs, and a new NVIDIA Dynamo Codelab focused on optimizing inference. developer live stream.
Over the past year, this community has become the go-to hub for AI builders using NVIDIA acceleration tools for data science and machine learning. The result is now production-ready. Search extension generation Measure observability of application and agent workloads on Google Kubernetes Engine (GKE).
These AI builders are also exploring new large-scale language models and experimenting with prototyping hybrid on-premises and cloud inference for real-world use cases like sports analytics and enterprise data pipelines.
Build with Google DeepMind’s Gemma, NVIDIA Nemotron, and Open Frameworks
NVIDIA and Google Cloud provide developers with learning resources and hands-on labs that combine NVIDIA libraries, open models, and tools with Google Cloud’s AI platform. This allows you to build optimized, production-ready AI applications faster.
For example, a developer can Data science and analytics Use or deploy the NVIDIA cuDF library for Google Colab Enterprise or Dataproc multi-agent application By combining Google DeepMind, gemma 4 Google Agent Development Kit with Google Cloud G4 VM with Google’s NVIDIA RTX PRO 6000 Blackwell GPU, NVIDIA Nemotron Open Model, Google Agent Development Kit cloud run Or use spot instances.
NVIDIA and Google Cloud work closely together across open frameworks, including: jacks So developers can get strong performance and a consistent experience while building, scaling, and productizing JAX workloads on NVIDIA AI infrastructure on Google Cloud, from single-GPU experiments to multi-rack deployments.
This effort also extends to the Google Cloud AI hypercomputer. max text The framework uses these JAX optimizations to efficiently train large models on NVIDIA GPUs.
Build on the same foundation, NVIDIA Dynamo on GKE helps developers optimize large-scale inference, including expert mixture models, and enables them to more efficiently deliver AI applications using NVIDIA accelerated infrastructure on Google Cloud.
To help developers put these capabilities into practice, a new learning path on Running and Scaling JAX on NVIDIA GPUs and a new NVIDIA Dynamo in GKE Inference Codelab will be available to members of the Google Cloud and NVIDIA developer communities next month.
Powering responsible AI with Google DeepMind’s SynthID and NVIDIA Cosmos
AI agents are increasingly built from systems of AI models, a combination of proprietary and open-source models, that reason, plan, and act on behalf of users.
Amid this change, trust and transparency will be fundamental so that developers and organizations understand how these systems work and what they produce.
Nvidia is first industry partner In conjunction with Google DeepMind Synth IDan AI watermarking technology that embeds robust digital watermarks directly into AI-generated content. This maintains the integrity of the output from your content. NVIDIA Cosmos World Foundation models are available at build.nvidia.com.
Cosmos models provide rich 3D perception and simulation capabilities for robots, autonomous machines, and other physical AI systems, and SynthID brings content transparency to the images and videos they rely on.
Together, they maintain the integrity of AI-generated content, allowing developers to more responsibly build and deploy agent applications across cloud, edge, and real-world environments.
Built on full-stack NVIDIA and Google Cloud platforms
This year, Google I/O is focused on new agent experiences and tools for developers. NVIDIA and Google Cloud are also focused on making sure builders have the infrastructure, software, and learning resources they need to get the most out of them.
Developers in the community building on NVIDIA and Google Cloud can extend the skills and tools they learn to easily move projects from prototypes to enterprise-grade workloads.
At Google Cloud Next, Google Cloud and NVIDIA extended their full-stack platform to help developers train, deploy, and operate agents on Google Cloud. This collaboration includes work on A5X instances powered by NVIDIA Vera Rubin, Google DeepMind Gemini models, and more, leveraged by leading AI labs and companies including OpenAI, Thinking Machine Labs, Schrodinger, Salesforce, Snap, and Crowdstrike. For more information, this blog.
participate Connect with other builders and stay up to date on new tools, developer events, and programs in the NVIDIA and Google Cloud developer communities.
