I’m a big ESP32 fan. They’re easy to develop on and very cheap to get into production with, if your project goes in that direction. For these reasons, they’re absolutely everywhere in the connected devices world. The new ones are lower power, and even have some machine learning/edge compute capabilities. But the hardware is going to depend on the systems engineering and where you’re planning on running the AI functions.
The easiest start you’ll have is to just find an open source project with similar capability and just piggyback on their system architecture
Red flags galore. There are very strict rules about how startups can raise money, how they can publicly disclose a raise, how shares are sold, etc. That in itself isn’t a red flag, Angel syndicates are a thing and billions have been raised through them. What’s concerning to me is your statement that you’ll inevitably be a unicorn is not something somebody would say who’s intimately familiar with VC/angel fundraising.
What is your key insight here? Are you a VC lawyer who drafted a new legal framework for startup fundraising? Do you have a giant Rolodex of startups and investors? Have you considered all of these potential roadblocks?