<!--IssueSummary start--> <details> <summary> Everyone can contribute. [Help move this issue forward](https://handbook.gitlab.com/handbook/marketing/developer-relations/contributor-success/community-contributors-workflows/#contributor-links) while earning points, leveling up and collecting rewards. </summary> - [Close this issue](https://contributors.gitlab.com/manage-issue?action=close&projectId=278964&issueIid=442896) </details> <!--IssueSummary end--> When an issue is created, developers already use resources like documentation, existing code, code comments, and community guidelines to formulate an implementation proposal for a possible solution. We could have an AI do the same. All the data in a repository: - existing code base with comments - Issues and related merge request - Documentation from within the repo and outside. (Usually, inline comments of dependencies are really comprehensive in what the code does) - Repo's wiki Could be used to train a model specialized to pick up data from Gitlab repos that would generate these proposals with a comment `danger-bot` style or generate an MR. Kind of like Gitlab Duo but with a deeper understanding of what needs to be achieved, and what has been done or could be done to achieve it.