Day 2 at Google Cloud Next: A marathon developer keynote
At Google Cloud, every day is Developer Day, but none so much as day 2 of Google Cloud Next, when we hold the developer keynote. This year’s topic? An in-depth look at Gemini Enterprise Agent Platform. This year’s theme? Planning a marathon for 10,000 participants through the Las Vegas Strip.
OK, let’s run with it.
Gemini Enterprise Agent Platform: A warm up
As the evolution of Vertex AI, **Agent Platform** “allows you to build autonomous agents that proactively help users — and complete tasks independently,” said Brad Calder, President, GCP and SRE. The platform does so with a whole suite of tools and capabilities to build, scale, govern, and optimize your agents.
Brad then passed the baton to keynote emcees Richard Seroter, Chief Evangelist, and Emma Twersky, Developer Relations Engineer, for an in-depth run-through of the agentic marathon simulator.
The system uses three main agents: A planner to determine routes; an evaluator that assesses routes based on business and community requirements; and a simulator that takes the route, adding actors and randomized behaviors to test the impact on the city,
This, Emma said, turns out to be “a great example of how agents can help us plan, simulate, and think about solving a really big challenge.”
First off the blocks: Building the agent
Mofi Rahman, Developer Relations Engineer, came onstage to demo how the **Agent Development Kit (ADK)**, Google Cloud remote **Model Context Protocol (MCP) servers**, and **Agent Runtime** provide the planner agent with the **Instructions**, **Skills**, and **Tools** it needs to improve the initial agent. By the end of the demo, the simulator had generated a route.
“The simulated route looks beautiful,” said Mofi. “Looks like the runners are going to get an amazing view of the entire Las Vegas Strip.”
An agent to evaluate the agent
Next, Ivan Nardini, Developer Relations Engineer, and Casey West, Architecture Advocate, showed us how to evaluate the agent, and build a UI for it. “We want to show you how to move from fragile, unpredictable agentic loops, to a rigorously evaluated network of experts that literally build their own UI,” Casey said.
They did so by deploying a separate model to judge the route, checking both deterministic (e.g., route length) and non-deterministic (e.g., community impact) criteria. For UI development, they showed off the **Agent-to-User Interface**, or **A2UI**, an open-source standard developed by Google that created an interface in a single shot. They also used the **Agent-to-Agent Protocol**, or **A2A**, and Agent Platform’s **Agent Registry**, to connect and see which agents are deployed.
“Think of Agent Registry as the DNS of your internet of agents,” Casey said.
Agents that never forget
Richard then mused about how to build agents that get better with time, that “take the learnings from the simulation and optimize for the next run.” Because the answer shouldn’t be to “cram raw text in every request we send back to our agents.”
To capture this learned knowledge, Agent Platform offers **Agent Platform Sessions** and **Memory Bank**, plus the ability to turn to tools like Spark or a database to retrieve more information, resulting in an even stronger simulator.
When agents go off course
Thus far, everything had gone swimmingly, but then Richard accidentally “broke” the simulator agent. That provided a perfect opportunity for Megan O’Keefe, Senior Staff Developer Advocate, to show off **Agent Interoperability** and **Gemini Cloud Assist**, and how to use them to debug agents at scale.
“With these autonomous agents, the production challenge isn’t just scaling the infrastructure, it’s managing the reasoning, the tool calls — all the places in the system where something can go wrong!” Megan said.
Megan used **Agent Runtime trace view** to see where the problem was, and using natural language, launched a **Cloud Assist Investigation** to explore logs and events, which pointed to a specific line of code as the offender. Megan then opened up her **Antigravity IDE** (powered by **Gemini 3**, and connected via MCP) to find the problem (an insufficiently run “event compaction” run) and to suggest a fix (add a token_threshold parameter to the event compaction config). She approved the fix and committed it to source, triggering a redeployment to **Agent Platform**. Problem solved!
Scaling the agents
To this point, all of the presenters had been showing off agent services running as **Cloud Run** services. Bobby Allen, Group Product Manager, then showed how to convert the apps to **Google Kubernetes Engine (GKE)**, which provides greater control, as well as to use a customized **Gemma 4** model, all by vibe coding in the Antigravity editor, which is connected to Cloud Assist. Along the way, Bobby also migrated the agents from **GCSFuse** to a high-performance **Lustre**file system.
Closely related to scaling is sharing — making agents available for the world to use and build on. Ines Envid, Senior Director, Product Management and Jason Davenport, Area Technical Lead, showed how to build no-code agents from the **Gemini Enterprise** app, and how to integrate them with other, “high-code” agents.
Shifting down
Last but not least, it was time to talk about security and governance. “Agents give users and other agents new ways to intentionally — or unintentionally — expose data and behavior in ways that we may not want,” mused Emma.
The standard response to that is to “shift left” — move testing, quality, and performance evaluation earlier in the development process — but for developers, that usually means more work, Richard said. “It’s not sustainable for developers to be responsible for all the layers of the stack,” he said. Instead, “we need to shift down.”
To help, there’s **Agent Identity** and **Agent Gateway**, demoed by Ankur Kotwal, head of Cloud Developer Relations. Ankur showed how Agent Gateway uses IAM policies to ensure agent actions are only accessible by approved sources, and how Agent Identity provides each agent with a unique and immutable credential. Then, **Agent Policies** can be configured to provide guardrails for the agents.
Yinon Costica, Co-Founder and VP of Product at Wiz, then went a step further and showed how Wiz can scan your agent code and infrastructure, and **Wiz Green Agent** can suggest root cause remediations.
“It’s a full architecture for security to easily understand what you built without you having to actually explain it,” Yinon said. Better yet, he also showed using this functionality from **Anthropic’s Claude Code with Opus.**“With Wiz, we want to enable your choice of tools and models to fix and prevent real risks,” he said.
The finish line
With this, the developer keynote came to an end. But for Google Cloud developers, it’s just the beginning, as the entire solution is available as source code in GitHub, and all the demos are available as Codelabs. Because when it comes to agentic development, these resources will really help you hit the ground running.
Posted in