iliashad
iliashad on Hacker News
Posted by
iliashad on Sunday November 01, 2026 @03:30PM
from the self-post dept.
TLDR: I indexed 37 hours and 38 minutes of my videos using the self-hosted Docker version of my project on a rented RTX 4090- Total compute time: 24 hours 13 minutes- Frame analysis took ~85% of the time- Heaviest plugins: TextDetection (especially on coding/screen recordings) and FaceRecognition (on podcasts)Full article here:...
I added the lower battery mode. haha.
Noted. Would like to show the app in a 15-minute call here (https://cal.com/iliashadad/orcah-studio-demo)
Yes, the Docker container is pretty different than the app, but it does share similar infrastructure with richer video scene data, direct integrations with Davinci Resolve, Final Cut Pro, and Adobe Premiere Pro, and a smart video search agent.
Thank you for your comment, much appreciate it
Posted by
iliashad on Monday October 05, 2026 @04:17PM
from the self-post dept.
Hey,I'm Ilias, the founder of Orcah Studio. I'm building a local-first desktop app (Mac OS available now) that can transcribe audio, recognize faces, detect objects, perform OCR, and describe video scenes.Plus, a local-first video agent that understands your prompt and finds the exact moment you're looking for. I fine-tune the agent based on an open-weight model (an 8-billion-parameter model)...
Posted by
iliashad on Thursday October 01, 2026 @05:59PM
from the show-hn dept.
Posted by
iliashad on Tuesday June 30, 2026 @07:28AM
from the show-hn dept.
TLDR: Following my recent blog post and Hacker News post (https://news.ycombinator.com/item?id=48528029). where I ran the desktop app on my M1 Max. This time, I’m using the self-hosted version, running in Docker, with an NVIDIA RTX 4090 (24 GB of VRAM).The content is also fundamentally more demanding: long podcast episodes with at least two faces in every frame, coding tutorials packed with...
TLDR: Following my recent blog post and Hacker News post (
https://news.ycombinator.com/item?id=48528029). where I ran the desktop app on my M1 Max. Today, I used Edit Mind (my source-available, local-first video indexing tool) to process my YouTube library, 87 videos, 37h of footage in under 24h on a rented RTX 4090 via Docker + CUDA.
Available on GitHub: github.com/IliasHad/edit-mind
Posted by
iliashad on Thursday June 25, 2026 @04:56AM
from the self-post dept.
I've been working on Edit Mind for a while now, and today I noticed Scott Hanselman starred it. He's VP of Developer Community at Microsoft. My project is a video knlowdge base that can help you index your videos using local machine learning models to let you search them using natural language or plain english.Github: https://github.com/iliashad/edit-mind
I'm not sure if it's possible or not, the goal of my project was to utilize local models and your local machine instead of uploading your videos to the cloud. there's couple of cloud services that offer video indexing very well, I could name a few like Tweleve Labs (I'm not affiliated with them but I did a presentation at one of their webinar).
That's good, also the Davinci resolve integration works with the Free and Studio version the same way.
You're welcome. As of now, the project narrow down your options instead of watching 3-4 videos (5 min each, as an example), you can watch 3 clips from them instead (1s to 2.5s each). It gives you only the video moments that you need to watch for review instead of watching the full videos.
That's great, would love to hear your feedback then
Can you check again ? I'm not sure why it's show a cloudflare error
Thank you! I'm using RAG, I have every video scene indexed individually in the vector database. When I'm asking the agent, it'll use an Ollama model to understand the request, use the available search tool (searching using transcription text, faces, visual, audio or combined) something like when you use Claude or Chat GPT it'll use the web search tool to find you info online. Then, I can filter out video scenes using the Ollama to better present accurate and unique video scene, then send those video results to Davinci Resolve using their API to create a video timeline using those video clips
That's really great, thank you!
Fair enough, what would like to see as an example video and I would make it.
For the dog barking videos, those are only the video scenes that I have a dog barking sound in the video.
I'll keep adding more prompts and example videos, keep an eye for that
Can you please elaborate more?
I agree with that, thank you for your feedback. Also, maybe you're not a video editor and you just wanna search your videos. The video editing integrations are optional and you have full control. You can switch between Adobe Premiere Pro, Final cut Pro or Davinci Resolve
Thank you, would like to know your use case for this kind of project and which prompt you want to genearte ?
Sure, I'm using (
https://huggingface.co/collections/Qwen/qwen25-vl) which can help me understand action like falling down, because I can provide for example 5 frames (down scaled to 720p) to understand what is happening in this part of the video
> Aha, it makes total sense. This number sounds much more reasonable than “669 GB”, since the actual total size of processed frames would be like 10-30 GB.
The reason why is “669 GB” is the total raw footage size when I'm doing the video processing, I downscaled each frame to 720p to make the video processing much faster and I don't need full original quality in order to get accurate results (as far as I know and experiment with).
> I’m just curious tho — is there any paying options that can accelerate this kind of process? Just spin up GPU instances?
For now, I found that NVIDIA GPU for example RTX 3060 with 12GB Vram was much faster than my M1 Max. (still working on optimizing for speed and accuracy).
That's great, I checked your article when it was in front page because someone mentioned my project in the comments.
Good job for the article and the project. That's great, yes local models are getting better and better