Hi everyone. We're in the process of evaluating Amazon Q Developer for our engineering team. Our main CI/CD pipeline (Jenkins) runs on-premises, and we manage our own GitLab instance. We're heavily invested in Terraform for AWS provisioning and have a mix of containerized and serverless workloads.
I'm trying to map out a practical integration path. The goal is to have Q assist with code suggestions and reviews within our existing developer workflows and pipeline stages, without a hard pivot to fully managed AWS services like CodeCatalyst.
My initial thoughts are around these points:
* **Local IDE Integration:** The VS Code extension seems straightforward for individual developers. Any gotchas with corporate proxies or internal CA certificates?
* **Pipeline Integration:** Could we use the Q API in a Jenkins pipeline stage for automated code analysis? For example, after a build succeeds, have Q review the changed code for security smells or suggest optimizations before merging.
* **Authentication & Secrets:** How are you managing AWS credentials for Q access from on-prem systems? I'm leaning towards using IAM roles for service accounts (if we can bridge that gap) or a dedicated pipeline user with tightly scoped policies.
Has anyone successfully wired Q into a similar hybrid environment? I'm particularly interested in:
* Any Jenkinsfile or pipeline script snippets you're willing to share.
* How you handled the "context" for Q, ensuring it's aware of our internal libraries and terraform modules.
* Whether the integration feels additive or if it requires significant pipeline re-architecture.
I can share our current pipeline structure if that helps frame the discussion.
Cloud cost nerd. No, I don't use Reserved Instances.