Professional Agentic Architect Preparation Details
The Professional Agentic Architect certification tests your ability to design, build, and govern autonomous AI agents on Google Cloud. This guide maps the beta exam’s domains, from Gemini Enterprise and the Agent Development Kit to Agent Identity and Agent Gateway, to official Google Cloud documentation. You can also explore more GCP certification study guides on the GCP to keep building your skills.
Google Cloud Professional Agentic Architect Materials
Section 1: Building agents using low-code tools (~13% of the exam)
1.1 Configuring agentic workflows and behavior using low-code tools. Considerations include:
Configuring state-based workflows (pages, transition routes, and event handlers) using Gemini Enterprise tools (e.g., Gemini Enterprise Agent Designer and Customer Experience Agent Studio [CX Agent Studio])
Customer Experience Agent Studio
Creating system instructions and in-console prompt templates (e.g., few-shot and chain-of-thought) to guide agent behavior (e.g., Agent Designer and CX Agent Studio)
Customer Experience Agent Studio
1.2 Connecting enterprise data to Gemini Enterprise. Considerations include:
Configuring agents to securely connect and query enterprise proprietary data sources (e.g., Gemini Enterprise and Agent Search)
Connect a third-party data source
Introduction to connectors and data stores
Use Agent Search as a retrieval backend using Vertex AI RAG Engine
Ingesting and processing unstructured multimodal data (e.g., videos, audio, and images) into the agentic workflow
Introduction to connectors and data stores
RAG Engine on Gemini Enterprise Agent Platform overview
Section 2: Using coding agents for application development (~17% of the exam)
2.1 Using coding agents effectively. Considerations include:
Configuring coding agents with Model Context Protocol (MCP) servers, custom skills, and access to tools (e.g., Antigravity and Claude Code on Google Cloud)
Authoring Google Antigravity Skills
Google Workspace MCP servers in Google Antigravity
Using coding agents in secure sandboxes (e.g., Google Kubernetes Engine [GKE], Cloud Workstations, and Antigravity)
Best practices for AI workload security on GKE
Using coding agents to refactor source code, optimize execution runtimes, and patch application-layer vulnerabilities
Scan and verify code vulnerabilities
Fix code vulnerabilities and manage diffs
2.2 Customizing coding agents for enterprise workflows. Considerations include:
Creating skills, plugins, extensions hooks, rules, and subagents using Antigravity
Authoring Google Antigravity Skills
Google Workspace MCP servers in Google Antigravity
Augmenting Antigravity with Agents CLI to build, scale, govern, and optimize deployed agents
Build an agent with ADK and Agents CLI
Section 3: Developing custom agents (~33% of the exam)
3.1 Designing and building agentic workflows in code. Considerations include:
Selecting and configuring the appropriate language model (e.g., large language model [LLM] vs. small language model [SLM], self-hosted vs. software as a service [SaaS], and open-source software [OSS] vs. proprietary LLM) considering cost, security, and agent architecture
Overview of Agent Development Kit
Building custom agents using open-source libraries (e.g., Agent Development Kit [ADK])
Overview of Agent Development Kit
Build an agent with ADK and Agents CLI
Configuring sessions and memory (e.g., Agent Platform Memory Bank and managed sessions)
Manage sessions with Agent Development Kit
Quickstart with Agent Development Kit
Configuring skills using Agents CLI (e.g., plugins and agent vs. human mode)
Build an agent with ADK and Agents CLI
3.2 Integrating enterprise domain knowledge. Considerations include:
Designing, configuring, and managing retrieval-augmented generation (RAG) pipelines and vector retrieval systems (e.g., embedding models, similarity scoring, and reranking) using appropriate services such as vector databases (e.g., Vector Search and Agent Retrieval)
RAG Engine on Gemini Enterprise Agent Platform overview
Overview of vector database choices
Configuring agent permissions (e.g., Agent Identity)
Using Google Cloud tools (e.g., Agent Registry, Google Cloud MCP Servers) to configure prebuilt and custom capabilities (e.g., custom integration layers for managed databases, API integrations, and MCP server that connects agents to third-party SaaS tools and remote servers)
Resolve endpoints and build orchestrators
3.3 Orchestrating and coordinating agentic workflows. Considerations include:
Orchestrating agents using agentic protocols (e.g., MCP and Agent2Agent [A2A])
Selecting and coordinating multiagent handoffs and workflows (e.g., parallel agents, sequential agents, and graph workflow) using Google Cloud tools (e.g., Agent Identity, Agent Registry, Agent Runtime, and agent policies)
Section 4: Evaluating and deploying agentic workflows (~22% of the exam)
4.1 Evaluating agents in development and in production. Considerations include:
Creating test sets for agent evaluation (e.g., golden data, prompts, and edge cases)
Creating continuous evaluation pipelines to assess an agent’s tool execution based on established success criteria
Continuous evaluation with Online Monitors
Determining the appropriate evaluation framework and tooling (e.g., ADK evaluation tooling (evalset), Agent Platform Gen AI evaluation service, and custom autoraters)
Evaluating an agentic system against a golden dataset to assess agent response and retrieval quality (e.g., using ADK)
Analyze evaluation results and failure clusters
4.2 Deploying and scaling production workloads. Considerations include:
Selecting optimal deployment runtime based on the use case, requirements, and cost (e.g., Agent Runtime, Cloud Run, and GKE)
Deploy AI agents with the ADK and Agent Platform API
Troubleshooting agent issues (e.g., drift, tool invocation latency, agent reasoning loops, and system failures)
Monitoring and optimizing agents for performance, reliability, and cost (e.g., identify logic errors, latency bottlenecks, and hallucinations)
Optimize and scale Agent Platform Runtime performance
Section 5: Securing and governing agentic workflows (~15% of the exam)
5.1 Configuring agent security and governance. Considerations include:
Implementing authentication and secure tool execution (e.g., agent-to-tool API calls using OAuth 2.0)
Authenticate to services using 3-legged OAuth with auth manager
Authenticate to services using 2-legged OAuth with auth manager
Authenticate to services using API key with auth manager
Configuring principal access boundary (PAB) policies using Agent Identity
Configuring Agent Gateway to monitor traffic and track agents
Designing and configuring agentic governance and policy enforcement (e.g., Agent Registry and Model Armor)
Configure Model Armor on a gateway
Semantic governance policies overview
5.2 Implementing secure agent behavior and execution. Considerations include:
Designing appropriate safety frameworks and guardrails (e.g., Agent Gateway, Model Armor, and human-in-the-loop [HITL])
Configure Model Armor on a gateway
Sensitive Data Protection overview
Configuring secure access to data and identity propagation (e.g., Agent Gateway and Agent Registry)
Authenticate to tools and resources
Wrapping Up Professional Agentic Architect
This guide covered all five domains of the Professional Agentic Architect exam guide, from low-code workflows in Gemini Enterprise to securing agents with Agent Identity and Agent Gateway. Work through the Agent Development Kit, RAG Engine, and Agent Runtime documentation hands-on before your beta exam window closes. You can also explore more GCP certification study guides on the GCP to keep building your skills. Have a question or tip? Leave a comment below.
Receive Updates on Professional Agentic Architect Exam
Want to be notified as soon as I post? Subscribe to the RSS feed / leave your email address in the subscribe section. Share the article to your social networks with the below links so it can benefit others.
