can anyone educate me on how viktor ai works on the backend? i use it for my business and it's really great, but im also wondering what's really going on under the hood because i figure maybe there's a way for me to make claude into something like viktor i'm sure a lot of it is based on jake's ICM structure, but what would it realistically take for me to have my own version of viktor but it's just claude?

today i stumbled upon a youtube video that talks about a tool called buttercut (it's free!) that helps edit talking head videos so much faster. and i thought i would share! it can edit multiple videos at the same time and do a solid first pass so you can get straight to making creative decisions and not edit out ums and ahs for 45 minutes. i made a couple of small changes to the repo (so i could use elevenlabs instead of whisper for transcription) but it works fantastic! i spent most of the day testing this workflow against some other tools, making some adjustments, and making sure it worked well for what we needed. here's the training video i sent my team: https://www.loom.com/share/8dd08df9f4114bf49bfcaa6df7102c72 and here's the video from the original creator explaining the tool: https://www.youtube.com/watch?v=FBkfr1yWf_s and here's the repo from the original creator if you want to try it yourself: https://github.com/barefootford/buttercut who this is for: - editors who want to work in their primary NLE and build customs workflows based on what they need - editors who have to have videos go through color grading passes, audio, and FX with detailed versioning without having to export between multiple tools to get a revision done - editors who may need to batch edit large amounts of videos to be exported into their NLE who this is not for: - someone who does not need to use davinci/premiere/fcp and have a pipeline where there is color/audio/fx/versioning - someone who films clips that require no color/advanced audio(you could just use descript in this case and that would work great!) problem this solved for me and my team: - we often get bogged down when we have to edit a bunch of talking head clips, and the first pass usually takes hours. - if we used descript to mitigate that time, it would require us to do color FIRST, then export into descript, and pray that the client wouldn't need any revisions on the color. (which can easily be 3 hours of rendering on a youtube video) - we've found descript's interface to be somewhat clunky for a lot of things we need. multitrack editing and complex audio don't seem to play well in descript (currently) - from what i currently understand, descript can't do this many isolated videos at the same time, it's usually expecting videos you upload to all be part of the same project, unless you explicitly tell it each piece of media belongs in a different composition, but even so, versioning for those quickly becomes disorganized if done in the same project - i am looking to have descript a bit more integrated into our workflow, but it doesn't fit every scenario. when it does work though, it works great. - no need to burn descript credits like crazy, i already have a claude max plan, so i can just utilize what i already have

yo what's up everybody. does anyone have any recommendations for how to use claude code in tandem with any other tool to automate the first cut of a talking head youtube video in davinci resolve? context: i run a video production company, and we’re trying to turn around long form youtube videos for clients while also turning around ads and short form deliverables for our other clients the reason i'm asking is because my editing team and i are all pretty bogged down with our workloads right now, and i’d like to relieve some of the pressure off of them to make their lives easier and increase output descript won't cut it because we do have to have a professional color grading workflow in the pipeline for this project (we’re using braw footage that i personally have to go in and denoise + grade, the render for that will take hours). if the client needs revisions, having all the files in davinci will make our lives so much easier. because it saves 3 hours of render time + re export into descript any suggestions?

This might be a dumb question but what exactly are the practical benefits of taking the folder based approach? Does it provide more efficient use of tokens, more consistent results? How can we demonstrate this so we can sell this idea to others? I tried creating a sample home page using the folder based approach versus standard claude.ai but couldn't really see a big difference.

whaddup y'all! i just got through pretty much everything from the free courses (except the archive ahha, that's next) and most of the vault too. this stuff is absolutely incredible!!! question for the group: how are some of y'all using these frameworks in your businesses lately? context on me: I run a video production company in LA, mainly working with another marketing agency where we turn around a ton of ads and assist with client campaign management, but i'm also working with music producers on their content to help them get more bookings. just curious to learn what y'all have got going on under the hood!