JetBrains/thinkrail — Vibe code with pi in a lightweight, real IDE that customises itself around the way you work — The Vibe You Need

What happened

JetBrains, a company known for coding tools, shared ThinkRail on GitHub as an incubator project. ThinkRail is built around the pi coding agent, which is an AI that writes and changes code for you. ThinkRail gives pi a full workspace with editors, file trees, terminals, and chats. The team says the next step is adding extension support, so ThinkRail can build the tools a user needs.

Key facts

• Each workspace is a separate git worktree, with its own branch and folder.
• Several pi chats can run at the same time in one workspace.
• Old or finished workspaces move to a shelf after their pull request merges or closes, or after a quiet week.
• Specs are plain SPEC.md files kept next to the code.
• The project is marked as a JetBrains incubator project.

Some key words

A git worktree is a second copy of a project folder that uses the same history. It lets an AI work on one task without changing your main files. A spec is a written note that says what the code should do. ThinkRail stores these notes so the AI remembers the goals of the project between chats. A skill is a set of instructions the AI can read and follow.

Why it matters

Many AI coding tools work in one place and forget what they did before. ThinkRail tries to keep project knowledge in files and let several AI tasks run side by side. This could help developers work faster and keep their main code safe. Because the project is still an incubator effort, its future is not yet certain.

What to watch

• Whether the planned extension support lands and how well it works.
• How the workspace and pull request tools perform in real projects.
• How the project grows beyond its early stage on GitHub.