Welcome to episode 300 of The Cloud Pod – where the forecast is always cloudy! According to the title, this week’s show is taking place inside of a Dr. Suess book, but don’t despair – we’re not going to make you eat green eggs and ham, but we WILL give you the low down on all things Vegas. Well, Google’s Next event which recently took place in Vegas anyway. Did you make any Next predictions? 

Titles we almost went with this week:

• ☁️This is the CLOUDPOD Episode 300
• 🍽️Tonight we dine in the Cloud
• 📚The Next Chapter
• Now in Preview: Episode 300

A big thanks to this week’s sponsor:

We’re sponsorless! Want to get your brand, company, or service in front of a very enthusiastic group of cloud news seekers? You’ve come to the right place! Send us an email or hit us up on our slack channel for more info. 

GCP

Pre-Next

02:35 Google shakes up Gemini leadership, Google Labs head taking the reins 

• There was a lot of Gemini news at Next – but we’ll get to all that. 
• In this particular case, there’s an employee shakeup. Sissie Hsiao is stepping down from leading the Google team, and is being replaced by Josh Woodward, who is currently leading the Google Labs. 

04:35 Filestore instance replication now available

• GCP says customers have been asking for help in meeting business and regulatory goals, and so they are releasing Filestore instance replication.
• This new feature offers an efficient replication point objective (RPO) that can reach 30 minutes for data change rates of 100 MB/sec.

05:16 Multi-Cluster Orchestrator for cross-region Kubernetes workloads

• The public preview of Multi-Cluster Orchestrator was recently announced.
• This lets platform and application teams optimize resource utilization, enhance application resilience, and accelerate innovation in complex, multi-cluster environments. 
• The need for effective multi-cluster management has become essential as organizations increasingly use Kubernetes to deploy and manage their applications; Challenges such as resource scarcity, ensuring high availability, and managing deployments across diverse environments create significant operational overhead.
• Multi-Cluster Orchestrator addresses these challenges by providing a centralized orchestration layer that abstracts away the complexities of underlying Kubernetes infrastructure matching workloads with capacity across regions.

06:26 GKE at 65,000 nodes: Evaluating performance for simulated mixed AI workloads

• Recently GKE announced it can now support up to 65,000 nodes (up from 15,000.) 
• Saint Carrie be with your CFO. 

09:15 How we built the new family of Gemini Robotics models 

• Worried about Skynet taking over? Now is the time to check out these articles so you can learn about our robot overlord’s weaknesses. 

09:58 Tuesday Night

Was anyone else weirded out by the scheduling? Did any listeners actually stay until the end on Friday? If you, we’d love to hear from you. 

13:30 The AI magic behind Sphere’s upcoming ‘The Wizard of Oz’ experience 

https://www.youtube.com/watch?v=f01dsTigSmw 

*Show note writer Heather is a curator at a museum that showcases Hollywood’s early history – and is VERY interested to see how the film world feels about this AI rebuilding of such a beloved classic. Some interesting discussions are definitely coming! 

21:11 Next Day 1 Keynote

Ironwood: The first Google TPU for the age of inference

• A seventh-generation Tensor Processing Unit (TPU) — our most performant and scalable custom AI accelerator to date, and the first designed specifically for inference.
• It is one of several new components of Google Cloud AI Hypercomputer architecture, which optimizes hardware and software together for the most demanding AI workloads. With Ironwood, developers can also leverage Google’s own Pathways software stack to reliably and easily harness the combined computing power of tens of thousands of Ironwood TPUs.

24:30 📢 Ryan – “So I was sort of surprised because they did spend a lot of time talking about inference and this chip handling inference concerns. I thought that was real. I mean, it’s just not the way that we’ve been talking about these custom AI chips in the past, right? It’s definitely been all about model training and building all these things. And the inference is more about running these very large models. And so there did seem to be a huge focus on performance and end user experience with AI development all the way through the conference.”

Google Workspace adds new AI tools to Docs, Sheets, Chat and more.

26:04 Google Agentspace enables the agent-driven enterprise

• Google says that in order to scale, businesses need AI-ready information ecosystems, and that’s why they’re launching Google Agentspace. 
• This product puts the latest Google foundation models, powerful agents, and actionable enterprise knowledge in the hands of employees. With Agentspace, employees and agents can find information from across their organization, synthesize and understand it with Gemini’s multimodal intelligence, and act on it with AI agents. 

27:24 📢 Ryan – “Well, so it SEEMS really cool, until you get through the hard edges…a lot of it really relies on your utilization of Chrome Enterprise Premium, and so that’s a whole workspace ecosystem that if you’re not bought into you’ve got a whole lot of heavy lifting to make that work.” 

32:18 New video, image, speech and music generative AI tools are coming to Vertex AI.

• Google’s new text to music AI offering, Lyria, is now in private preview. 
  • This means customers can generate complete, production-ready assets starting with a text prompt.
• Veo 2 has new editing and camera controls features available in preview with allowlist that help enterprise customers refine and repurpose video content with precision.
• Chirp 3 now includes Instant Custom Voice, a new way to create custom voices with just 10 seconds of audio input.
  • Hopefully soon we’ll be ready to present our new and improved Jonathan with the help of Veo 2. 
• Imagen 3 has improved image generation and inpainting capabilities for reconstructing missing or damaged portions of an image and is making object removal edits even higher quality.

AI Hypercomputer updates from Google Cloud Next 25

36:46 Agent Development Kit

• This open-source framework simplifies the process of building sophisticated multi-agent systems while maintaining precise control over agent behavior. Agent Development Kit supports the Model Context Protocol (MCP) which provides a unified way for AI models to access and interact with various data sources and tools, rather than requiring custom integrations for each. 

43:30 Agent 2 Agent (A2A)    

• We’re proud to be the first hyperscaler to create an open Agent2Agent protocol to help enterprises support multi-agent ecosystems, so agents can communicate with each other, regardless of the underlying framework or model. 
• More than 50 partners, including Accenture, Box, Deloitte, Salesforce, SAP, ServiceNow, and TCS are actively contributing to defining this protocol, representing a shared vision of multi-agent systems. 

Google Unified Security

• This solution brings together our visibility, threat detection, AI powered security operations, continuous virtual red-teaming, the most trusted enterprise browser, and Mandiant expertise — in one converged security solution running on a planet-scale data fabric. 

Cloud WAN

Meta Llama 4

Other Day 1 items:

We’re introducing a new way to analyze geospatial data.

48:20 Next Day 2 Keynote

52:16 Vertex AI Agent Engine

• Vertex AI Agent Engine (formerly known as LangChain on Vertex AI or Vertex AI Reasoning Engine) is a fully managed Google Cloud service enabling developers to deploy, manage, and scale AI agents in production. Agent Engine handles the infrastructure to scale agents in production so you can focus on creating intelligent and impactful applications. Vertex AI Agent Engine offers:
  • Fully managed: Deploy and scale agents with a managed runtime that provides robust security features including VPC-SC compliance and comprehensive end-to-end management capabilities. Gain CRUD access to multi-agent applications that use Google Cloud Trace (supporting OpenTelemetry) for performance monitoring and tracing. To learn more, see deploy an agent.
  • Quality and evaluation: Ensure agent quality with the integrated Gen AI Evaluation service.
  • Simplified development: Vertex AI Agent Engine abstracts away low-level tasks such as application server development and configuration of authentication and IAM, allowing you to focus on the unique capabilities of your agent, such as its behavior, tools, and model parameters. Furthermore, your agents can use any of the models and tools, such as function calling, in Vertex AI.
  • Framework agnostic: Enjoy flexibility when deploying agents that you build using different python frameworks including Agent Development Kit, LangGraph, Langchain, AG2, and LlamaIndex. If you already have an existing agent, you can adapt it to run on Vertex AI Agent Engine using the custom template in our SDK. Otherwise, you can develop an agent from scratch using one of the framework-specific templates we provide.

Data analytics innovations at Next’25 | Google Cloud Blog

• Google introduced advancements in data analytics, emphasizing the integration of AI and intelligent agents to enhance data accessibility and decision-making. 
• Key innovations include specialized agents for various roles, AI-assisted data science workflows, and an autonomous data foundation in BigQuery that supports unstructured data and open formats like Iceberg.

Gemini Code Assist in IDEs

Software Engineering Agents – now in preview! 

Google Cloud Next 2025 Wrap Up:

• An application-centric, AI-powered cloud | Google Cloud Blog
• Chrome Expands AI-Powered Enterprise Search and Enterprise Browser Protections | Google Cloud Blog
• Accelerate your analytics with new Bigtable SQL capabilities | Google Cloud Blog
• Migrating SQL server databases is now easier with Gemini. Here’s how | Google Cloud Blog
• Announcing intelligent unified governance in BigQuery | Google Cloud Blog
• AlloyDB AI drives innovation for application developers      

1:07:22  Google Next Predictions

• • Ryan – No Points. Sad. 
    • Responsible AI, in Console/Service/SDK to enable and/or visualize your responsible AI creation or usage
    • Endpoint Security Tools (Crowdstrike, Patch Management/Vulnerability)
    • Won’t be announcing anything new service announcements just enhancements for AI/Gemini/Etc.
  • Justin – 1 Point 
    • AI Agents specialized for Devops, K8, Devops capability
    • Next Generation of TPU GPU’s optimized Optimized Multi-modal
    • Unification or Major Enhancement of Anthos & GKE Enterprise 
  • Matt – All the points
    • Green AI
    • 3 not-AI specific keynotes
      • Cloud WAN
      • Hyperdisk Exapools
      • Next Gen Customer Engagement Suite
    • AI security thing that is not Endpoint. More Guardrails. 
      • Google Unified Security
  • Honorable Mentions – nada 
    • Industry verticalization for AI LLM Models. Fine Tuning Marketplace or special model for specific industry/use case
    • Personal Assistant for Workspace productivity 
    • Multicloud tooling
  • Number of times AI or ML said on stage

• Opening Keynote 38 times

• Developer Keynote: 63 Times

• Total: 101

• • Matt: 52 
  • Justin: 97
  • Ryan: 1

Closing

And that is the week in the cloud! Visit our website, the home of the Cloud Pod where you can join our newsletter, slack team, send feedback or ask questions at theCloud Pod.net or tweet at us with hashtag #theCloudPod