AI & LLMs · Guide · AI & Prompt Tools
How to Use Amazon Q Developer
Installing Q Developer in VS Code/JetBrains, inline completions, agent tasks, security scans, AWS integration.
Amazon Q Developer is AWS’s coding assistant — the successor to CodeWhisperer — offering inline completions, agent-style tasks, and deep AWS service awareness.
Advertisement
Rebranded from CodeWhisperer in 2024, Amazon Q Developer keeps the IDE completions engine and adds a chat agent, automated code upgrades (Java 8/11 to 17/21), and security scans tuned for AWS SDK patterns. It’s Amazon’s direct answer to Copilot and Cursor.
What it is
Q Developer runs as an IDE plugin (VS Code, JetBrains, Visual Studio, Eclipse) and as a CLI. It covers three surfaces: inline suggestions as you type, a side-panel chat that can read your workspace, and agent tasks (/dev, /transform, /test) that produce multi-file pull requests. It reads your AWS account when you opt in, so it can answer questions like “why is this Lambda timing out” with real config.
Install / sign up
# VS Code code --install-extension AmazonWebServices.amazon-q-vscode # JetBrains: Settings > Plugins > search "Amazon Q" # CLI (macOS) brew install amazon-q # Sign in with AWS Builder ID (free tier) or IAM Identity Center (Pro)
First session
After signing in, open a file and start typing — completions appear in grey, tab to accept. Open the Q panel to chat, or run /dev to kick off an agent task that produces a diff.
$ q chat > /dev add a DynamoDB table and wire it into the orders Lambda # Q generates CDK changes, Lambda edits, and tests # review the diff, then /accept
Everyday workflows
- 1. Use /transform to modernise a Java 8 service to Java 17 with Spring Boot upgrades applied automatically.
- 2. Run /review to get a security-focused scan that flags IAM over-permissions and SDK misuse.
- 3. Ask the chat panel AWS questions grounded in your account — billing, CloudFormation drift, or service limits.
Gotchas and tips
The free Builder ID tier covers generous monthly inline suggestions and a handful of agent invocations; Pro ($19/user/month) lifts the limits and adds customisation on your private repos. Customisations require an S3 bucket of reference code and a one-time training job.
Q Developer won’t suggest code that matches public repos verbatim when reference tracking is on — leave that setting enabled for license hygiene. For non-AWS projects the agent still works, but you lose the account-aware superpowers.
Who it’s for
AWS-heavy teams, Java shops planning an LTS upgrade, and anyone who wants Copilot-class completions wired directly into their cloud console.
Advertisement