@KevinMagnan has a decision model like Jev pick the retake winners from the transcript while editing TikToks.

Product note, text only, quoting @petergyang. Since January, a decision model like @typesafeai Jev picks the retake winners from the transcript. Each cut is spliced as audio and re-transcribed before render so words are not clipped. Apple Vision checks that no overlay touches the face. A Sunday retro reads the numbers and feeds lessons back into the skills once they hold twice. The post also lists pieces taken from @petergyang, including the frame-pinned review page. For people editing short video with a decision model.