Recently I started to look into the code review data our team generated through Azure DevOps Pull Requests (PR’s). While it was easy to get basic metrics like
- Number of PRs created
- Number of PRs reviewed
- Number of PRs voted
- Number of review threads created
It is difficult to analyze the content of the code review comments to address questions like “who are the leading contributors that provide thorough, constructive feedback in the PR threads?”, “what are the most frequently discussed coding issue types?” “Based on code review feedback, what are the recommended training programs for individual developers?”
I’d like to explore whether GenAI can help answer these questions with their reasoning and summarization capabilities.
Recently I started to look into the code review data our team generated through Azure DevOps Pull Requests (PR’s). While it was easy to get basic metrics like
It is difficult to analyze the content of the code review comments to address questions like “who are the leading contributors that provide thorough, constructive feedback in the PR threads?”, “what are the most frequently discussed coding issue types?” “Based on code review feedback, what are the recommended training programs for individual developers?”
I’d like to explore whether GenAI can help answer these questions with their reasoning and summarization capabilities.