Help GitHub developers label issues more consistently and act on them faster by using an LLM (Gemini) to predict "UI" vs. "Other" for real issues, then trace each labeled issue back to its PR(s) and commit(s). The pipeline turns a repo’s raw history into linked, reviewable artifacts (commit to PR to issue) so teams can compare human labels with model predictions, spot gaps, and streamline triage and fixes.
Questions can be answered:
- How much code ships through PRs?
- What share of changes close issues?
- Which releases have more UI-related fixes?
Log in or sign up for Devpost to join the conversation.