◆ For solo YouTube creators
Vanpelt Studio puts a writer, a story editor, an assistant editor, a packaging designer and an analyst behind your channel. AI included, nothing to set up, and your footage never leaves your machine.
Twenty minutes with the person who uses it most. No deck, and I'll tell you if it won't help you.
What it is
Vanpelt Studio is a desktop app that holds the story you meant to tell and the footage you actually got, then reads one against the other.
01Write the listabout an hour
Every line you need to say and every shot you need to get, written down before you roll instead of re-decided twice afterwards.
02Drop the footage inminutes, not evenings
Point at the card. Everything is transcribed and indexed locally, so a moment is something you search for, not something you scrub for.
03See what's missingthe same evening
Your list, graded against what the card actually holds. The holes surface while a pickup is still twenty minutes, not on day twelve when they cost you the week.
The 40-second version
What you're doing alone
A studio would put five people on this before anyone touched a timeline: a writer to decide what the video is, an assistant editor to watch every clip back, a story editor to fix the order, and someone to package it. You are all four of them, plus whoever was supposed to notice the shot you never got, and that is why three weeks go by while only two days of it were editing. With the team behind you, the same video takes three days - two of which are the editing you keep.
You, alone≈ 3 weeks
With Vanpelt Studiothe same jobs, three days of your time
Your job, unchanged: the same two days of editing in both.The four jobs you were doing for free. Read down a column to see what each one costs you now.
What this is, and what it isn't
AI video tools race to cut your footage for you. Vanpelt Studio asks the question they skip: do you have the video? Every cut still belongs to you.
Hey, I have all this footage from my vacations. Can we make something interesting with it?
Sure. Here are all the clips you could take to make a funky vlog story.
The night after the shoot
“When I sat down to edit the video, I realised while watching the footage how much I was struggling… I was having a hard time knowing what I wanted the video to look like.”
“I have so much filmed, but I don’t know where to start.”
“It seems to take forever. It’s almost like I have to look at the footage for the same amount of time I’m riding.”
“You can easily lose two hours “getting familiar” with your footage before making a single cut.”
That struggle has a cause: you find out whether you have a video after you’ve committed 20 hours to editing it. Vanpelt Studio moves that answer to the night after the shoot - while a missing piece is still a 20-minute pickup.
Real words from public creator posts and interviews - describing the problem, not our product.
Local-first
Your clips, the transcription and the frame extraction all stay on your disk. What crosses the wire is text you could read yourself (a transcript line, a beat, a shot description), and it goes to whichever model you picked. Never the video, never the audio.
Six ways to connect, chosen explicitly: your existing Claude or ChatGPT subscription, your own Anthropic or OpenAI key, ten open models through OpenRouter (DeepSeek, Qwen, Llama, Mistral, Gemma, Kimi, GLM), or the AI bundled with Studio. No silent switching between them.
The rest of the card
Every other tool slices your video into clips your audience already watched. This one mines the footage you didn't use: the tangents, the asides, the mishaps sitting in the hours that never made the cut. A candidate that's just a highlight of the main video is thrown out on purpose: it has to work for someone who will never watch the long one.
Want to build a section around one thing? Ask for it by name and get every clip it appears in, matched on what was said and what's on screen, not on filenames you never wrote. Collect the hits into a bin and hand that to your editor instead of a four-hour card.
Made with it
These are from my own channel - not customer testimonials. Vanpelt Studio exists because I needed it for these, and each one went through the spine, the footage check and the packaging pass you've just read about. They're public: go watch one and judge the tool by the result.
Yes, we made an infomercial · 46s
Talk to a human first
I built Vanpelt Studio for my own channel and I'm still its heaviest user. If you're weighing it up, I'd rather spend twenty minutes showing you what it does with real footage than have you guess from a landing page. No deck, no script. Bring your channel, your footage situation and your worst editing week.
And if it isn't going to help you, I'll say so on the call. I'd rather lose the sale than the reputation.
Founding buyers get an onboarding call anyway. This one happens before you decide.
No. Vanpelt Studio is local-first: transcription runs on your computer, and only lightweight derived text (transcripts, your beat sheet) is sent to the AI model you chose. Your raw footage never leaves your disk.
Yes. Vanpelt Studio plans, organizes and assembles the structure of your video, then hands a ready-to-cut project to your editor (like DaVinci Resolve). You keep full control of the final edit.
macOS or Windows, and a way to reach an AI model, which you pick in Settings. If you already pay for Claude or ChatGPT, the app can use that subscription directly, so there is nothing extra to buy. You can also bring your own Anthropic or OpenAI API key, or use the AI bundled with Studio and skip the setup entirely. Whisper and ffmpeg run locally and are bundled, so there is nothing else to install.
You choose, and the choice is explicit: your existing Claude or ChatGPT subscription, your own Anthropic or OpenAI key, or the AI bundled with Studio. There is no silent failover between them, so a signed-out or rate-limited backend tells you that instead of quietly spending money somewhere else. With your own key you pay the provider directly for exactly what you use, with nothing marked up and no middleman between your footage-derived text and the model.
Vanpelt Studio is built for long-form: the 8, 15, 30-minute videos where structure is the hard part and where a copilot saves you the most time.
Three days of work, not three days of wall clock, and it comes out of a specific place: the deciding, the searching, and the rework. You still shoot, you still cut, and a video with a hard shoot or a heavy edit takes longer. What goes away is re-deciding the structure twice, watching everything back to remember what you have, and finding the missing shot on day 12. If your videos are already tightly planned and your footage is already organised, you will save less. The tool is worth the most when the mess is the problem.
One subscription · everything included
Writer, story editor, assistant editor, packaging and analyst - AI included, nothing else to set up. Your footage never leaves your machine.
Rather bring your own AI? Same app, your Claude/ChatGPT or API key:
