Edit Mind

Orcah Studio picks one job and does it well: turning a folder of untagged footage into something you can ask questions of. The indexing is genuinely local — after the first video pulls down the models, transcription, face grouping, object counting and scene descriptions are computed on the Mac, and the pitch that you can turn the internet off and keep working is the whole reason this product exists. That matters most for unreleased client masters and NDA-bound documentary work, where uploading rushes is a non-starter. The search surface is the strong part. Speaker-labelled transcription means a made-up quote can be written as "find me where @Ilias explains MoviePy" and land on a second. Visual Match takes a frame you already have — a look, a colour, a composition — and returns ranked similar shots with timecodes. Scene descriptions turn a jump off a boat into something searchable as exactly that, and the telemetry layer reads what GoPro, Strava and cinema sidecars already wrote — GPS, T-stop, focal length, colour science, LUT, focus distance, lens serial — and makes it filterable. The handoff into Final Cut Pro and DaVinci Resolve (Free and Studio) closes the loop so you do not export and re-import by hand. The honest weaknesses. It is a companion, not an editor: there is no timeline to cut on, so it sits beside your NLE. It is Apple Silicon only — an Intel Mac or a Windows machine is not a supported target today, and Windows is a waitlist. There is nothing here for teams: the offer is a single lifetime license, and the evidence names no shared libraries, seats or permissions. It is sold as a public beta, so buy it expecting rough edges rather than a finished release. Adobe Premiere Pro is not a standard integration — it is request access only — so if Premiere is your edit home, confirm before you buy. Where it fits: editors and documentary teams with hundreds of hours of unsearchable rushes, studios that cannot put masters in the cloud, and self-hosters who already run a NAS. Where it does not: Windows and Linux shops, teams that need shared libraries, and anyone who wants search and cut in one window. Note the free alternative — the legacy self-hosted Docker project (Whisper, YOLO, DeepFace, ChromaDB) is still on GitHub, free and source-available; at $249 you are paying for the native Mac interface, the agent and the direct NLE handoff rather than the search itself.