Udemy
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
Professional Agentic Architect: GCP Exam Prep
New
2 students

Professional Agentic Architect: GCP Exam Prep

Master agent design, ADK orchestration, evaluation, and security to prepare for Google's new Agentic Architect exam.
Created byAseem Mankotia
Last updated 10/2026
English

What you'll learn

  • Design agent solutions using Vertex AI Agent Builder, Agentspace, and Conversational Agents
  • Build custom multi-agent systems with the Agent Development Kit (ADK)
  • Select and apply multi-agent orchestration patterns (sequential, parallel, hierarchical, routing)
  • Implement tool/function calling, memory, and session/state management in agents
  • Ground agents using Vertex AI Search, BigQuery, and custom RAG pipelines
  • Evaluate agent trajectory, groundedness, and safety using the Vertex AI evaluation service
  • Deploy, version, and observe agents on Vertex AI Agent Engine
  • Apply least-privilege IAM, VPC Service Controls, and secrets management to agents

Course content

13 sections • 12 lectures • 2h 59m total length
  • Course Overview and the Agentic Architect Exam Blueprint13:07

Requirements

  • 3+ years of hands-on Google Cloud experience (recommended by Google)
  • 1+ years building agentic or generative AI solutions on Google Cloud
  • Working knowledge of Python
  • Familiarity with Vertex AI and generative AI concepts
  • Completion of the Google Cloud Generative AI Leader or Professional ML Engineer course recommended

Description

This course contains the use of artificial intelligence.


Google Cloud's Professional Agentic Architect certification is the newest and most advanced credential in the generative AI track, validating the ability to design, build, evaluate, and secure production AI agents and multi-agent systems on Vertex AI. This course, created by Aseem Mankotia, delivers exam-focused preparation built specifically for the five official exam domains published in Google's beta exam guide, giving architects, AI engineers, and developers a structured path from the Generative AI Leader and Professional ML Engineer courses into true agentic architecture.


Across twelve chapters you will work through Vertex AI Agent Builder, Agentspace and Conversational Agents for low-code agent design, Gemini Code Assist for coding-agent workflows, the Agent Development Kit (ADK) and Agent Engine for custom multi-agent systems, the Vertex AI evaluation service for trajectory and groundedness testing, and the security and governance controls (IAM, VPC Service Controls, Model Armor, and prompt-injection defenses) that production agent systems require. Every chapter pairs scenario-driven theory with a console or ADK-based hands-on lab, and quiz questions mirror the case-study style of the real exam's multiple choice and multiple select items.


The course closes with a full timed simulation covering roughly eighty questions across all five domains, plus a three-hour time-management strategy for exam day. Because this credential was authored during Google's beta window ahead of general availability on November 2, 2026, candidates should confirm the current exam guide on the official Google Cloud certification page before scheduling their exam. This is exam-focused preparation for the agentic-AI era, built to help experienced Google Cloud practitioners master every domain of a genuinely new architect role.


AI content disclosure: This course was produced with the assistance of artificial intelligence tools. Lecture narration is AI-voice generated, and lecture scripts, slides, and practice questions were drafted with AI assistance, then reviewed and curated by the instructor for technical accuracy and alignment with the official Professional Agentic Architect exam guide.

Who this course is for:

  • Cloud architects, AI engineers, and application developers who design and deploy production AI agents and multi-agent systems on Google Cloud. Candidates should have solid Google Cloud and Vertex AI experience plus at least a year building agentic solutions, and comfort with Python and generative AI concepts. A strong next step up from the Google Cloud Generative AI Leader and Professional ML Engineer courses.