Studio
In developmentA senior video editor who learns how you cut, and who starts the moment your filmmaker drops the material.
Agente · cuts the videoHere is how it works today. Your filmmaker shoots, takes the cards home, dumps everything onto a laptop, sorts the material, watches it back to see what is usable, and then starts editing. Saturday goes out on Wednesday, by which point nobody cares.
What changes here is not the speed. It is where the work starts: the moment the material lands in the folder, someone is already cutting it. And that someone does not go to sleep.
It learns how you cut
This is what separates it from any automatic clipper, and it is worth understanding before anything else.
You give it your back catalogue. Everything you have published so far. Your judgement comes out of that: how long your cuts run, where you come in, where you get out, which shot opens, what never makes it. It is not a template that treats a club in Ibiza and a club in Tenerife the same.
And it keeps learning from what you tell it. More of this, less of that. Every note you give goes into the judgement, so month six looks more like you than month one did. That is the difference between a tool you configure and an editor you train.
You can also hand it a brief and ask for a piece built from that, instead of waiting to see what comes out of the session.
What it does with the material
It finds the tempo and the drops. It cuts on the music. It grades with LUTs. It assembles, adds subtitles and leaves every piece in the format of the platform it is going to. And it keeps everything indexed, so you can ask for “the moment that track dropped” and it comes up.
By the next morning, instead of a folder of rushes, your social team finds pieces ready to post to Instagram and TikTok.
And it watches what works outside, not only inside
A system that only learns from your own content gets better and better at what you already did. Studio also reads what is running across the platforms — formats, pacing, what the algorithm is pushing this week — and ideas come out of that which were never in your archive.
Two more things it settles, that nobody sells you
Delivery control. You set out what your filmmaker owes you — so many videos, so many drops, so many stills — and the system checks whether it arrived. It stops being an argument and becomes a figure.
A reading of floor energy, session by session and artist by artist. That is hard data for booking: you stop deciding who to rebook on a hunch.
Where it is today
In alpha. The pipeline runs locally on real material from Papagayo Tenerife, session by session, and it works end to end; what is missing is the product around it. Internally it is called Anaga.
And publishing carries human approval by default. Turning that off is your decision, not the opening offer.
What it answers
Why did last year's best Friday sell out?
You can reconstruct that event and repeat what worked.
How long does Saturday's video take to go live?
The edit starts while the room is still open.
How many pieces did you put out last month, and how many did you want to?
What you put out and what you wanted to, side by side.
And what answers them
- Data Science
What it does not do yet
- Publishing automatically without human approval
- Filming for you — your team shoots it, we pick it up from the folder
We say it here so we do not have to take it back in a meeting.