I see some existing posts about a new user interface.
This is desperately needed. The current system is almost unusable by mere mortals as it stands now.
Are you planning to implement a native iOS app for the re-design or just tweaking the horrible web view that is "wrapped" for mobile?
I hope you don't take this the wrong way, but why do ex-Googler's build products with such terrible UI? I tried to use another ML/AI video system from ex-Googler's (you may know them, it is called Camio) and amongst other problems, the UI for that is terrible too. Yes it's cloud based, different architecture and pricing/business model, but the UI is even less usable.
OTOH, the alerts and recognition in Camect are working really nicely so far, but alerts are useless if you can't get to the relevant footage, search footage easily, and otherwise navigate through the system.
I know there is huge difference between building back-end and front-end systems. It you can't build a world-class UI, have you considered partnering with others that have the skill and expertise?
Personally, I would see more benefit if you could integrate with an existing great UI and be able to put all your time into improving the ML/AI, adding facial recognition, and maybe going deeper into "video search" like Camio. Nobody else has anything like this at this kind of price point and you should be leveraging that not re-inventing the UI that other teams have already built.
At a minimum, your engineers should do a deep dive on Ring, Nest, and Unifi Protect to see very capable systems. Also, look at Hikvision, and the other junk sold in big box stores and commercial legacy systems for examples of what NOT to do. Apple's HomeKit Video and HomeKit Secure Video is also a great example of terrible UI (which is very surprising to see from Apple)