NVIDIA and Google Cloud announced their joint developer community has grown to over 100,000 members at this year's Google I/O conference. The community provides developers with learning paths, hands-on labs and events focused on building AI applications using NVIDIA and Google Cloud technologies.

Launched at Google I/O last year, the community serves developers, data scientists and machine learning engineers working with NVIDIA and Google Cloud AI tools. New additions rolling out this year include a learning path for using the JAX library on NVIDIA GPUs, an NVIDIA Dynamo codelab for inference optimizations, and monthly developer livestreams.

The community has facilitated development of production-ready applications including retrieval-augmented generation applications on Google Kubernetes Engine and observability tools for agent workloads. Members have also worked on large language model research and hybrid cloud-premises inference for use cases in sports analytics and enterprise data pipelines.

NVIDIA and Google Cloud are providing resources that combine NVIDIA libraries and tools with Google Cloud's AI platform. Developers can use NVIDIA cuDF library in Google Colab Enterprise or Dataproc, and deploy applications using Google DeepMind's Gemma 4 models and NVIDIA Nemotron models with Google Cloud G4 VMs powered by NVIDIA RTX PRO 6000 Blackwell GPUs.

The companies are collaborating on Google DeepMind's SynthID watermarking technology, which embeds digital watermarks in AI-generated content. NVIDIA became the first industry partner to work with Google DeepMind on SynthID for NVIDIA Cosmos world foundation models available on build.nvidia.com.

A new learning path on JAX and NVIDIA GPUs and an NVIDIA Dynamo codelab will become available next month for community members, according to the companies' joint announcement.