JetBrains Air is an agentic development environment that orchestrates multiple AI coding agents like Codex, Claude Agent, Gemini CLI, and Junie. It enables parallel task execution in isolated environments, provides full oversight, and integrates code-aware review workflows for efficient software development.
Freemium
How to use JetBrains Air?
Define a coding task by referencing files or commits. Launch multiple AI agents (Codex, Claude, Gemini, Junie) to work on tasks concurrently in isolated Docker containers or Git worktrees. Monitor their progress in real-time, switch context to provide input, and finally review and commit the generated changes using intelligent code navigation.
JetBrains Air 's Core Features
Run multiple AI coding agents (Codex, Claude Agent, Gemini CLI, Junie) in parallel without conflicts, dramatically speeding up development cycles.
Delegate complex work by breaking it into subtasks and assigning the most suitable AI agent for each job, ensuring optimal tool usage.
Maintain complete oversight with enhanced visibility into agent activities, allowing you to check progress and course-correct tasks as needed.
Improve code review workflows by testing different agents, tools, and processes in a single, unified environment for better outcomes.
Define tasks with precision using code-aware context like symbols, classes, methods, and commits, ensuring agents understand the exact requirements.
Execute tasks in fully isolated environments using Docker, Git worktrees, or (soon) cloud setups, preventing conflicts and maintaining project integrity.
Manage concurrent agent sessions and complex workflows efficiently using dedicated workspaces that preserve history and avoid branch collisions.
JetBrains Air 's Use Cases
A software developer can rapidly prototype a new feature by having different AI agents simultaneously work on the backend API, frontend UI, and database schema.
A team lead can delegate bug fixing across a large codebase by splitting the work among specialized agents, reviewing all changes in one consolidated view.
A developer learning a new programming language can use Air to generate example projects and explain code, with multiple agents providing different perspectives.
An open-source contributor can automate repetitive tasks like dependency updates or code refactoring across multiple branches using parallel agent execution.
A solo entrepreneur can accelerate product development by using agents to handle boilerplate code, documentation, and testing concurrently.
