i built my own solution - mostly custom (aside from hardware), shut off from the internet. It has an option to sync to <cloud provider> but isnt required - have a fleet of (3) cameras at my house that i fully control and operate - have been running fully autonomous for the last month.
I did it because RING is too easy to setup and give your valuable footage and life to. I havent marketed it yet
Im building a script for my own dev to spin up droplets on DigitalOcean to run high performing LLMs like H100 - this is all prompted through open web ui (ollama) . I find the models my 12GB GPU will tolerate have too many hallucinations - its ok for diffing for example or basic scripts, but anything somewhat complicated I need 60GB+ GPU which isnt feasible for me to buy..
I did it because RING is too easy to setup and give your valuable footage and life to. I havent marketed it yet