Amazon Q Developer

Amazon Q Developer sits somewhere between GitHub Copilot and a full autonomous agent. The agentic mode is where it earns the premium tier: you describe a feature or refactor in plain language, and it generates a multi-file implementation plan, writes the code, runs tests, and iterates on failures without hand-holding. I've used it to upgrade a Spring Boot project from Java 11 to 17 across twenty-three files. It caught deprecation warnings, rewrote incompatible stream operations, and adjusted test fixtures. The whole job took twelve minutes of agent time versus what would have been an afternoon of grep and manual edits. The AWS integration is the obvious differentiator. If you work in Lambda, ECS, or any AWS service with an SDK, Q Developer pulls live context from your account. It suggests IAM policies that actually match your resource ARNs, writes CloudFormation that references existing VPCs, and spots security issues like overly permissive S3 bucket policies during code review. This is legitimately useful if you live in AWS. If you don't, most of that value evaporates. Failure modes: it struggles with legacy codebases that lack clear module boundaries. I pointed it at a Django monolith with circular imports and it generated a plan that would have broken half the app. The agent also can't access external documentation beyond what AWS provides, so if you're working with niche libraries, it falls back to generic suggestions. The CLI agent is underwhelming compared to the IDE extensions; it feels like an afterthought. Compared to Cursor or Windsurf, Q Developer is less fluent at general refactoring but significantly better at AWS-specific tasks. Cursor would need multiple prompts and copy-pasted IAM docs to achieve what Q does in one agentic run. For polyglot teams or non-AWS infrastructure, Cursor is the better bet. For teams already paying for AWS support, the Q Developer tier bundled with CodeWhisperer starts to make financial sense around five users.