Tool or Technology
TT0160

Mastering Google Antigravity

Advanced Agentic Development, Multi-Agent Engineering & Automation

Antigravity
Duration
32 Hours
Level
Advanced
Delivery
Instructor-Led
Format
Capability Development

Programme Objectives

  • Develop advanced expertise in using Google Antigravity for autonomous software engineering, research, debugging, testing, code modification, browser interaction, and complex knowledge work.
  • Design multi-agent engineering workflows using parallel subagents, Projects, multiple repositories, Git worktrees, specialised agents, and structured task delegation.
  • Extend Antigravity through Agent Skills, Plugins, MCP servers, custom tools, Rules, JSON Hooks, browser tools, and reusable engineering standards.
  • Build custom autonomous agents and application-level agentic workflows using the Antigravity SDK and CLI.
  • Establish enterprise controls for permissions, sandboxing, tool approvals, browser access, IAM, network restrictions, data residency, auditability, and human oversight.

Technology covered

Google Antigravity 2.0Antigravity CLIAntigravity SDKAntigravity IDEIDE ExtensionsGemini 3.8 FlashAgentsParallel SubagentsProjectsGit WorktreesArtifactsBrowser AgentChrome DevTools MCPAgent SkillsRulesPluginsMCP ServersJSON HooksScheduled TasksRemote ControlIntegrated TerminalVCS PanelCustom Tools & Enterprise Controls

Who should attend

  • Software Developers & Application Engineers
  • Full-Stack Developers
  • Frontend & Backend Developers
  • AI / Generative AI Engineers
  • Software Architects
  • Solution Architects
  • Technical Leads & Engineering Leads
  • DevOps & Platform Engineers
  • Site Reliability Engineers
  • QA & Test Automation Engineers
  • Application Modernisation Teams
  • Developer Productivity Teams
  • AI Agent Developers
  • Engineering Managers
  • Enterprise AI & Innovation Teams

Prerequisites & Participant Readiness

  • Working knowledge of at least one programming language
  • Familiarity with Git, repositories, branches, commits, and pull requests
  • Basic understanding of software development, debugging, testing, and code-review practices
  • Command-line and terminal familiarity is recommended
  • Basic understanding of Generative AI and agentic-AI concepts
  • Familiarity with APIs, JSON, Python, and external developer tools is beneficial for SDK and MCP modules
  • Basic knowledge of cloud and security concepts is helpful for enterprise modules
  • No previous Antigravity course attendance is required
  • Availability of enterprise, IDE Extension, model, Remote Control, and other advanced capabilities may depend on plan, platform, organisation configuration, and rollout status

TOC Modules

Concepts
  • Understanding Antigravity as an agent-first development platform
  • Understanding Antigravity 2.0, CLI, SDK, IDE, and IDE Extension surfaces
  • Understanding agents, tools, knowledge, context, and autonomous execution
  • Understanding Gemini-powered local agent architecture
  • Differentiating conversational coding assistance from autonomous engineering
Practical activities
  • Exploring an Antigravity 2.0 Project
  • Running a representative repository-analysis task
  • Comparing interactive development and delegated Agent workflows
  • Mapping Requirement → Agent → Tools → Code → Verification → Artifact

Scenarios

Multi-Agent Requirement-to-Production Development

Product Requirement → Antigravity Project → Implementation Plan → Main Agent → Parallel Subagents → Git Worktrees → Code Implementation → Browser Validation → Test Agent → Artifact Review → Human Approval → Merge

Participants design an advanced software-development workflow in which multiple specialised Agents simultaneously handle implementation, testing, research, and validation while Git worktrees and Artifact reviews keep development isolated and controlled.

Autonomous Engineering Maintenance System

Scheduled Task → Antigravity Agent → Repository Analysis → Skill → MCP / Development Tools → Issue Identification → Subagent Investigation → Proposed Fix → Automated Tests → Human Approval → Git Update

Participants build a governed recurring engineering workflow for activities such as CI investigation, dependency maintenance, technical-debt identification, documentation updates, or repository health checks while retaining explicit human control over consequential changes.

Related programmes

Continue with programmes from the same capability area.

Take the next step

Ready to make this programme work for your team?

Customise modules, duration and business scenarios for your team.

Instructor-ledVirtualHybrid

Designed around your roles, tools and real workflows.