Skip to main content
AI-105Axalon Original

Master the AI Coding Tool Built for Enterprise

Open-source, governed, and Google Cloud native — learn Gemini CLI the enterprise way

Taught by Dan King, 4x Google Trainer of the Year

1 dayinstructor-ledIntermediate
Gemini CLIEnterpriseAI CodingGoogle CloudMCP

Public class policy: Classes run with a minimum of 6 participants. If minimum enrollment isn't reached, you'll be notified 7 days before with options to transfer or receive a full refund.

Enterprise-First Approach

Start with the governance story. Learn Policy Engine, Trusted Folders, and compliance hooks before writing a single line of AI-generated code — so you can champion adoption with confidence.

Beyond Code Completion

Gemini CLI is a full agentic coding tool with sub-agents, multimodal input, and MCP integrations. This is not autocomplete — it is an AI teammate that understands your codebase.

Free and Open Source

No subscription required. Gemini CLI is Apache 2.0 with a generous free tier. Enterprise teams upgrade to Vertex AI for governance, not to unlock features.

Hands-On Every Module

6 labs across the day. Configure security, scaffold a microservice, build an MCP server, and deploy to Cloud Run. You leave with real skills, not slide decks.

Overview

Learn to adopt Gemini CLI across your engineering team with enterprise-grade security, governance, and Google Cloud integration. This hands-on course takes an enterprise-first approach — starting with the Policy Engine, Trusted Folders, and compliance hooks before diving into agentic coding with sub-agents, multimodal context, MCP integrations, and Cloud Run deployment.

What You'll Learn

  • Evaluate Gemini CLI against other AI coding tools for enterprise use
  • Configure enterprise security controls using Policy Engine, Trusted Folders, and compliance hooks
  • Use agentic coding with sub-agents and approval gates for safe code generation
  • Create hierarchical GEMINI.md context with multimodal input for team-wide standards
  • Build custom MCP servers and discover Agent Skills for extending capabilities
  • Deploy applications to Google Cloud Run and integrate with GCP services
  • Design an AI coding tool adoption strategy with measurable success criteria

Who Should Attend

Enterprise developers evaluating or adopting AI coding tools

Products Covered

Gemini CLICloud ShellCloud RunVertex AIArtifact Registry

Course Modules

1

The Enterprise Case for AI Coding Tools

Topics

  • AI coding tools landscape — Gemini CLI vs Claude Code vs GitHub Copilot CLI vs Cursor
  • Open-source advantage — Apache 2.0, inspectable code, no vendor lock-in
  • Cost model — free tier vs Vertex AI for governed usage
  • Security architecture — sandboxing, Policy Engine, Trusted Folders, .geminiignore
  • Data flow and privacy — /privacy command, telemetry controls
  • Enterprise readiness checklist

Learning Outcomes

  • Compare AI coding tools across security, cost, and capability dimensions
  • Articulate Gemini CLI's enterprise value proposition to stakeholders

Activities

Lab: Hands-on orientation — install Gemini CLI, authenticate with Vertex AI, run /privacy, /stats, /policies list, document findings for security team
2

Governance & Security Controls

Topics

  • Policy Engine — allow/deny/ask_user actions with regex patterns, priority-based rule resolution
  • Trusted Folders — restricting capabilities in untrusted directories, /permissions command
  • Hooks for compliance — BeforeTool for policy enforcement, AfterTool for audit logging
  • .geminiignore — excluding sensitive files from context
  • Network controls — proxy routing, egress restrictions
  • Vertex AI backend — governed model access with audit trails, VPC-SC compatibility
  • Team configuration — version-controlled settings.json, wrapper scripts

Learning Outcomes

  • Configure Policy Engine rules with deny-by-default posture
  • Set up compliance hooks for audit logging
  • Manage Trusted Folders and file exclusions

Activities

Lab: Configure locked-down installation — Policy Engine rules, BeforeTool compliance hook, .geminiignore, Trusted Folders, Vertex AI backend
3

Core Capabilities & Agentic Coding

Topics

  • Installation and authentication — API key vs Vertex AI, Cloud Shell quick-start, /init
  • The ReAct loop — reason, plan, act with approval gates
  • Code generation — single-file and multi-file from prompts
  • Refactoring — architectural changes across a codebase
  • Built-in tools — run_shell_command, Google Search grounding, PTY support, file system
  • Sub-agents — codebase_investigator for mapping architectural dependencies
  • Plan review — reviewing and editing agent plans before execution

Learning Outcomes

  • Use agentic coding patterns with approval gates for safe code generation
  • Leverage codebase_investigator for enterprise-scale refactoring

Activities

Lab: Scaffold a FastAPI microservice from requirements, use codebase_investigator to analyze structure, refactor to add authentication with approval gates
4

Context Management & Multimodal Input

Topics

  • Hierarchical GEMINI.md — home to project to module to component layering, /init for auto-generation
  • Team standards as context — coding conventions, framework patterns, security policies
  • File imports — @file.md syntax for modular context composition
  • Multimodal input — @diagram.png for architecture diagrams, @spec.pdf for design specs, audio/video
  • Memory system — /memory show, /memory add, /memory refresh, save_memory tool
  • Conductor extension — official extension for Context-Driven Development

Learning Outcomes

  • Design hierarchical context files for team-wide coding standards
  • Use multimodal input to provide visual and document context

Activities

Lab: Create GEMINI.md hierarchy for a team project, feed architecture diagram as multimodal context, verify with /memory show
5

MCP & Skills — Extending Gemini CLI

Topics

  • Model Context Protocol — tools, resources, prompts
  • Official Google MCP servers — Cloud Run, Google Cloud Services, Workspace
  • Building custom MCP servers — FastMCP for Python, STDIO transport
  • Authentication patterns — API keys, OAuth, service account impersonation
  • Enterprise MCP governance — approval and audit strategy
  • Agent Skills — /skills for discovery, activate_skill for specialized personas

Learning Outcomes

  • Build and integrate custom MCP servers for internal systems
  • Discover and activate Agent Skills for specialized tasks

Activities

Lab: Complete a FastMCP starter template for DORA engineering metrics, configure in Gemini CLI, query deploy frequency, browse Skills with /skills
6

Google Cloud Integration Deep Dive

Topics

  • Cloud Shell integration — zero-install, pre-authenticated
  • Cloud Run deployment — /deploy for one-line deployments, Artifact Registry
  • /setup-github — GitHub Actions for automated issue triage and PR reviews
  • Vertex AI integration — GOOGLE_GENAI_USE_VERTEXAI=true, service account auth, VPC-SC
  • GCP MCP servers — Cloud Run MCP, Google Cloud Services MCP
  • Cost tracking — /stats for token usage, Vertex AI billing dashboard

Learning Outcomes

  • Deploy applications to Cloud Run using Gemini CLI
  • Integrate with GCP services and monitor usage costs

Activities

Lab: Deploy Lab 3 microservice to Cloud Run via /deploy, inspect logs with Cloud Run MCP, review token usage with /stats
7

Enterprise Adoption Playbook

Topics

  • Adoption strategy — pilot, evaluate, expand pattern
  • Measuring impact — /stats for token usage, productivity metrics, developer satisfaction
  • Common objections — security, cost, accuracy — and how to address them
  • Template artifacts — GEMINI.md templates, Policy Engine configs, hook library, MCP starters
  • Ongoing learning — community resources, release cadence, Conductor for team knowledge

Learning Outcomes

  • Design an AI coding tool adoption strategy with measurable success criteria
  • Address common enterprise objections to AI coding tools

Activities

Discussion: Design adoption plan for your organization with template handouts

Get This Training

No public classes currently scheduled. Express interest below or request private training.

Request Private Session

Course Details

Course Code
AI-105
Duration
1 day
Format
instructor-led
Level
Intermediate
Hands-on Labs
6
Modules
7

What Makes This Workshop Different

Policy Engine & Security

Configure allow/deny/ask_user rules with regex patterns. Set up Trusted Folders, .geminiignore, and compliance hooks for enterprise-grade control.

Agentic Coding with Sub-Agents

Use the ReAct loop with approval gates. Leverage codebase_investigator for architectural analysis across large enterprise codebases.

Multimodal Context

Feed architecture diagrams, PDF specs, and audio into Gemini CLI as context. Generate code from visual designs and written specifications.

Custom MCP Servers

Build FastMCP servers that connect Gemini CLI to your internal APIs. Query engineering metrics, internal documentation, and team data through natural language.

Google Cloud Integration

Deploy to Cloud Run with a single command. Set up Vertex AI backend, GitHub Actions integration, and cost monitoring.

Adoption Playbook

Walk away with templates, Policy Engine configs, hook libraries, and a structured pilot-evaluate-expand strategy for your organisation.

I needed to convince our security team that AI coding tools could be used safely. This workshop gave me the technical proof — Policy Engine configs, audit hooks, Trusted Folders. I walked in sceptical and left with a deployment plan.

James Rivera

Senior Developer, Financial Services Firm

Frequently Asked Questions

What do I need to know before attending?
Basic terminal/CLI experience and familiarity with Google Cloud. No prior experience with AI coding tools is required — we start from installation and build up.
Is this just a Gemini CLI tutorial?
No. This is an enterprise adoption workshop. We cover governance, security, team configuration, and adoption strategy alongside the technical skills. You leave ready to champion the tool, not just use it.
How does Gemini CLI compare to Claude Code or GitHub Copilot?
Module 1 covers an honest comparison across all major AI coding tools. Gemini CLI differentiates on open-source transparency, free tier, 1M token context, and native Google Cloud integration.
Do I need my own GCP project?
No. We provide sandboxed lab environments with everything pre-configured. You just bring a laptop with a modern browser.
What if my company has strict security requirements?
That is exactly what Module 2 addresses. You will configure Policy Engine rules, Trusted Folders, network controls, and compliance hooks — the same controls your security team will want to see.
Do I get a certificate?
Yes. You receive an Axalon Training certificate of completion. The skills covered align with Google Cloud Professional Developer and Cloud Engineer certifications.

Give Your Team AI Coding Tools They Can Actually Use at Work

Enterprise-governed, open-source, and Google Cloud native — adopt Gemini CLI with security built in from day one

Free