AI BootCamps (7 Courses)
Course Category
Course Category
Google Rating
Professionals Upskilled
Live Cohorts Delivered
Enterprise Teams Trained
This 12-hour live Enterprise AI Agents on Google Cloud program takes you from a new Google Cloud project to a multi-agent system running in production on Gemini Enterprise Agent Platform, the platform that replaced Vertex AI in 2026. You deploy a machine learning model, enable Gemini and Model Garden foundation models, build a retrieval-augmented generation (RAG) application over your own documents, build single and multi-agent systems, and operate them with tracing, guardrails and cost controls.
Every module ends in a working lab. You build agents in Python with the Agent Development Kit (ADK), connect tools through the Model Context Protocol (MCP), link agents with the Agent2Agent (A2A) protocol, and deploy to Agent Runtime.
The course covers the custom-agent and deployment skills in Google Cloud's Professional Agentic Architect exam, a practical foundation for that certification.
12 Hours. 6 Modules. 10 Hands-On Labs. - Live instructor-led training using real Google Cloud resources.
Build the Full AI Application Path - Deploy ML models and foundation models, then build LLM applications, RAG solutions and AI agents.
Build Agents with ADK - Create code-first AI agents in Python using Google's Agent Development Kit, MCP tools and memory.
Build Multi-Agent Systems - Create coordinator and specialist agents with shared state, human approval and A2A communication.
Deploy and Operate AI Agents - Run agents on Agent Runtime with tracing, evaluation, Model Armor, identity and cost controls.
Build Toward Professional Agentic Architect - Develop the hands-on foundation for the Google Cloud Professional Agentic Architect certification.
Google Cloud essentials for AI builders
Agent Platform architecture
Lab 1: environment setup
Outcome:
Lab 2a: deploy an ML model
Lab 2b: use foundation models
Customize and control
Outcome:
Prompting and structured output
Lab 3a: RAG with RAG Engine
Lab 3b: guardrails and evaluation
Outcome:
Agent design
Lab 4a: build an agent with ADK
Lab 4b: tools with MCP
Outcome:
Orchestration patterns
Lab 5: build a multi-agent system
Agent-to-agent communication
Outcome:
Lab 6a: deploy agents
Lab 6b: observability and evaluation
Governance, cost and wrap-up
Outcome:
The biggest value was going beyond building agents. I learned how to deploy, monitor and operate multi-agent systems on Google Cloud, which is much closer to the work I handle in enterprise architecture.
The hands-on ADK labs made the concepts practical. Building agents with MCP, connecting them through A2A and deploying them on Agent Runtime gave me experience I could directly apply to real projects.
This course gave me a clear path from LLM applications to production-grade multi-agent systems. The focus on tracing, guardrails and cost controls made it especially relevant for enterprise workloads.
It covers custom agents and deployment, the largest parts of the exam. Plan for the exam-prep add-on to cover low-code tools and coding agents.
Google moved Vertex AI into Gemini Enterprise Agent Platform in 2026. The course uses the current names and points out the old ones.
You need agent fundamentals. The GenAI and Agentic AI BootCamp provides them, and equivalent hands-on experience also qualifies.
No. The ML model is pre-trained and the foundation models are built by their vendors. You deploy or enable them, then build the LLM application and agents.
Labs use small models and short-lived resources, and Module 1 sets a budget alert. Delete the resources after the final demo to stop charges.
Yes. Corporate cohorts are available with flexible scheduling.