Peter Duffy

I’m Peter Duffy, and with my co-founder Dalila I build Rooster, a transcript-first video editor for the Mac. My background is in marketing technology - years spent turning complicated requirements into simple, product-led experiences. Rooster grew out of a frustration I kept running into: editing a talking video meant hunting through a crowded timeline when what I actually wanted to change was the words. And the tools that let you edit by text sent your footage to someone else’s servers and charged by the minute for it. Rooster puts recording, transcription, text-based editing, captions and export in one native Mac app. You record or import a video, Rooster transcribes it, and you edit the video by editing the transcript - delete a sentence and the clip goes with it. It cuts the ums, retakes and dead air, cleans up your voice, adds captions, titles and lower thirds, and exports a finished MP4 ready to post. Everything runs on your own Mac. Transcription and the AI tools use on-device models, so your media never leaves the machine unless you choose otherwise, and there’s no media-minute allowance or AI-credit meter. Your bill stays the same however much you make. We believe your Mac is already an AI computer. Most everyday AI jobs - transcribing a recording, cutting the filler words - don’t need a data centre. Doing them locally is cheaper, faster and more private, and leaves the cloud for the jobs a Mac genuinely can’t handle. We also think AI should take on the slow parts of production without putting words in your mouth or a stand-in in your place. Rooster is for real people, in their own voice, saying what they mean. Rooster is small, independent and a member of NVIDIA Inception. Support messages are read by the two of us, so if something is unclear or harder than it should be, tell us. Try it free at rooster.cam.