Uploaded July 2026 | Updated September 2026, 2 weeks ago
Learn how to simply and securely scale production-grade AI agents in minutes – entirely on our Agent Platform. Our end-to-end platform for enterprise developers to accelerate agentic development from concept to production without compromising on reliability or governance
Building an agent is the easy part. The real challenge is the lifecycle: managing deployments, governance, evaluation, and publishing across fragmented tools.
In this session, you’ll learn how to turn your existing coding agent into a platform expert. First, you’ll learn how to build and deploy custom workflows using the Agent Development Kit (ADK) and Agents CLI. Then, you’ll discover how to scale execution with the agent runtime and memory bank, govern securely via the Agent Gateway and Agent Registry, and continuously optimize performance using agent observability and evaluation.
You’ll learn how to:
- Build no code, low code or high code agents directly in Gemini Enterprise Agent Platform
- Scaffold an ADK agent and build deterministic tools from plain-English prompts, then run it locally
- Scale and Deploy to Agent Runtime and add Sessions and Memory Bank so the agent keeps context across sessions
- Give the agent an identity and screen untrusted input for prompt injection
- Govern your agents securely with Agent Gateway and Agent Registry
Grade the agent on grounding and hallucination, then optimize the prompt behind a no-regression gate
- Publish the finished agent into Gemini Enterprise so it is available to your entire organization
Resources
Gemini Enterprise Agent Platform Overview - docs.cloud.google.com/gemini-enterprise-agent-platform/overview
Agents CLI Documentation - google.github.io/agents-cli
ADK Framework Guide - https://adk.dev/
#GoogleCloud #AgentDevelopment #ADK #GenerativeAI #CodingAgent #CloudComputing #AgentGovernance
Speakers: Shubham Saboo Lavi Nigam
Products Mentioned:, ADK, Agents CLI, Vertex Agent Platform, Model Armor, Gemini Enterprise
Learn how to simply and securely scale production-grade AI agents in minutes – entirely on our Agent Platform. Our end-to-end platform for enterprise developers to accelerate agentic development from concept to production without compromising on reliability or governance
Building an agent is the easy part. The real challenge is the lifecycle: managing deployments, governance, evaluation, and publishing across fragmented tools.
In this session, you’ll learn how to turn your existing coding agent into a platform expert. First, you’ll learn how to build and deploy custom workflows using the Agent Development Kit (ADK) and Agents CLI. Then, you’ll discover how to scale execution with the agent runtime and memory bank, govern securely via the Agent Gateway and Agent Registry, and continuously optimize performance using agent observability and evaluation.
You’ll learn how to:
- Build no code, low code or high code agents directly in Gemini Enterprise Agent Platform
- Scaffold an ADK agent and build deterministic tools from plain-English prompts, then run it locally
- Scale and Deploy to Agent Runtime and add Sessions and Memory Bank so the agent keeps context across sessions
- Give the agent an identity and screen untrusted input for prompt injection
- Govern your agents securely with Agent Gateway and Agent Registry
Grade the agent on grounding and hallucination, then optimize the prompt behind a no-regression gate
- Publish the finished agent into Gemini Enterprise so it is available to your entire organization
Resources
Gemini Enterprise Agent Platform Overview - docs.cloud.google.com/gemini-enterprise-agent-platform/overview
Agents CLI Documentation - google.github.io/agents-cli
ADK Framework Guide - https://adk.dev/
#GoogleCloud #AgentDevelopment #ADK #GenerativeAI #CodingAgent #CloudComputing #AgentGovernance
Speakers: Shubham Saboo Lavi Nigam
Products Mentioned:, ADK, Agents CLI, Vertex Agent Platform, Model Armor, Gemini Enterprise
