Turn customer feedback into tested pull requests
In OpenAI's own launch example, a Dot watches feedback for recurring requests, scopes small fixes, builds and tests them, and returns PRs with demo videos.
Done with Dots
What they did
The source gives this as a short scenario for a developer working on an app. The dot runs on its own cloud computer with connected apps. It watches customer feedback for recurring requests, scopes smaller improvements and bugfixes, builds and tests them, and hands back complete PRs with videos of the changes. The developer reviews them while building the next major feature.
How it went
No results or numbers are given. It is an illustrative example from the launch announcement, not a measured case. The page says dots can still make mistakes, so consequential work needs review.
Worth knowing
Dots are rolling out on Pro, Business Premium and Enterprise plans in eligible markets. Tasks a dot starts in Codex count toward your usage limits.
Try it yourself with Dots
Watch [feedback source, e.g. support inbox, forum, or survey sheet] for requests that recur at least [number] times. For each one, scope a small fix in [repo name], build it, run the tests, and open a draft pull request with a short demo video and a summary of the feedback behind it. Stop after [number] PRs and show me the list before merging or pushing anything beyond drafts.
Source
Read the original