# Ole is taking you through this one.
# Free, and it runs right here. No signup.
$ learn JetBrains AI Assistant
Learn JetBrains AI Assistant in 5 minutes.
JetBrains AI Assistant brings AI into IntelliJ, PyCharm, and the rest of the IDE family, working through the same static analysis the IDEs are known for. The lesson drills project-aware chat, completion habits, and AI shortcuts for tests, docs, and commits.
# 5 minutes, no install
# interactive, not a video
# Lesson not loading? Open it in a new tab. It is the same free lesson either way.
Chat beside the code, or actions inside it?
The assistant shows up in two forms: a chat window that knows your project, and small AI actions woven into the editing you already do. Most people find one and never meet the other.
Project-aware questions
A chat that can reference your codebase, so answers point at your actual classes and files rather than generic examples. Use it for understanding, planning, and multi-step help.
Completion, two flavors
Fast local completion for single lines and cloud-backed suggestions for larger blocks, layered onto the IDE's own type-aware completion. Knowing which layer produced a suggestion tells you how much to trust it.
Built-in chores
Generate a commit message from your staged diff, draft documentation for a function, create tests, explain a cryptic error. Small actions, sitting exactly where the chore occurs.
What you will do in the lesson
You answer and try things. Nobody demonstrates at you for five minutes.
- Ask about the project, not about code in general. You'll phrase questions so the assistant uses codebase context, and compare the answers against generic ones. Specific in, specific out.
- Rank three completion sources. IDE static completion, local AI, cloud AI: you'll look at suggestions from each and reason about when each source deserves trust.
- Generate tests, then grade them. You'll have tests written for a function and evaluate what they cover, what they miss, and which asserts are theater.
- Let the diff write the commit message. You'll generate a commit message from staged changes and practice the edit that turns an accurate summary into a useful one.
- Explain an error where it happened. You'll use the assistant on a failing run, right in the IDE, and see how far in-context explanation gets you before a search engine would.
Before you start
How is this different from installing a generic AI plugin in my IDE?
Integration depth. The assistant sits on top of JetBrains' static analysis, so it knows types, structure, and usages the way the IDE does, and its actions appear inside existing workflows like the commit dialog and quick-fix menus. Generic plugins bolt a chat window on; this is wired into the machinery.
Do I need IntelliJ or PyCharm running for this lesson?
No. The lesson runs in the frame on this page, free, with nothing to install. What you practice here maps directly onto the assistant when you next open a JetBrains IDE.
Can I keep completions local for privacy?
Partly. The fast single-line completion runs on your machine, while larger suggestions and chat use cloud models. If your constraint is strict, the lesson's habit of knowing which layer produced a suggestion doubles as knowing which suggestions ever left your laptop.
GitHub Copilot also runs in JetBrains IDEs. Why use this instead?
Copilot in a JetBrains IDE is a strong completion engine visiting someone else's house. The JetBrains assistant is the house: it shares the IDE's code understanding and puts actions into refactoring, commits, and error handling. Completion quality is comparable; integration is where they part. Many developers simply try both for a week in their real project.
Other tools people learn next
Same format, same five minutes, still free.
Learners from these organizations are already learning on 
