Polyglot in Both Code and Cloud | I Speak All Dialects of Expensive
IT Services and IT Consulting
IT Services and IT Consulting
Mar 2020 – Jul 2020
May 2019 – Oct 2019
Mar 2011 – Jun 2011
Mar 2023 – Mar 2023
Jan 2020
Dec 2023
Mar 2024 – Mar 2024
Jul 2024 – Jul 2024
May 2024 – May 2024
Mar 2024 – Mar 2024
Jun 2024 – Jun 2024
Jan 2024 – Jan 2024
Mar 2024 – Mar 2024
Apr 2024 – Apr 2024
Mar 2024 – Mar 2024
Post
Working on my media platform and wound up making something cool. This is a motion animation system with scripting. It's called Pulse and it has its own editor as well. Whats it used for? Initially its for making animated visuals on audio stations so I don't have to spend so much time and money on GPU's. It's got the added benefit of being very low bandwidth. Check it out and let me know your thoughts. -C
Post
I ran out of coffee, and today I'm doing a metric ton of Machine Learning. I hope the GPU's hold up until I hit Costco.
Post
Working on a full system for building AI directed Videos with locked characters. I've already generated a lot of stuff, most not worth looking at but helping tune and build the capability. My end goal is a personal one. I want to be able to make customized videos on demand. I like reading and one day would love for the tooling to be commoditized to the point where I could make my Series. The technology isn't there yet. Render time being the #1 issue. But if you think at scale, I could render faster than you can watch if i had enough compute. My thought process is more about bringing capability close to home to be prepared for what's next. Not my goal, but I do like the idea of a tuned custom feed, or lofi music for work on demand. So this is my exploration in that.
Post
I give random Homelab stuff to this one eager young guy at work. Every day I hear of some new dev misadventure he's undertaking. If you've got some spare servers or things you've stepped away from. Gift them to the younger generation and watch them take off with it. It's literally the most satisfying thing ever.
Post
Toying with Video generation right now I'm breaking down workflows so that I can have a tight lock on generation. Below are initial failures at training Loras that know what I look like. I'd say 70% there. My dataset was 15 images, the new one is over 900. I thought I'd share what training an Avatar looks like from the backend. You do a lot of training, and play with different models and see where things land. The tinkering part is fun, though it's a lot of waiting. Basically you feed it lots of reference images in different settings and iterate a couple thousand steps of training. Then you can use the LoRA on a model to force your character style/appearance. This is a 2 step process. as I'll be getting the base LoRA setup then applying modifications to turn it into a cartoon. The idea is to experiment with a educational short form video instructor.
Post
Toying with Video generation right now I'm breaking down workflows so that I can have a tight lock on generation. Below are initial failures at training Loras that know what I look like. I'd say 70% there. My dataset was 15 images, the new one is over 900. I thought I'd share what training an Avatar looks like from the backend. You do a lot of training, and play with different models and see where things land. The tinkering part is fun, though it's a lot of waiting. Basically you feed it lots of reference images in different settings and iterate a couple thousand steps of training. Then you can use the LoRA on a model to force your character style/appearance. This is a 2 step process. as I'll be getting the base LoRA setup then applying modifications to turn it into a cartoon. The idea is to experiment with a educational short form video instructor.