Getting the chance to talk to Peri soon, but here's my take..

- Backwards compatible C64/Amiga (I'm aware they don't own the Amiga Trademark itself), not PC/Emulation based (FPGA, SOC, Classic/Original Chips) - New capabilities, e.g. faster clockrates, co-processors/accelerators, support for more colors - keep ~240p as internal resolution, with analog i/o outputs + 4k internal upscaler for hdmi output - Introduce new tile-based mode, but keep bitmap/fb based ones - EPROM based cartridges (e.g. no filesystem/sd-like media) but still upgradable using cartridge-bound signed/encrypted updates files (via USB or via system cartridge slot (like Mod Retro/Chromatic does) - Wifi (mostly for multiplayer, maybe a builtin marketplace, not browsing the web!) - Exclusive game titles + license best modern 2d games - Effectively built a 2D powerhouse (think Neo Geo 2.0), in favour over 3d (people can still do that in software if cpu/2d capabilities are strong enough (see PICO-8 examples that do vr racing, cad editor and what not)

To achieve this, prioritize hiring engineers that have the expertise to build this new world over hiring managers or pc-based/x68/modern computing engineers. I'd recommend getting a strong, full-time technical lead like Gideon Zweijtzer, Furrtek, or one of the capable FPGA core developers from Mist/Mister.

Basically don't become another toy/mini console rebirth, but rather becoming the new cool kid on the block.

The original Commodore was known for a super cheap computer that made it practical for people to have one at home, as well as innovative graphics for the time.

I don't know how that might translate into modern times when everyone has a home computer (a telephone at least). Maybe a computer targeted to pre-teens?

Make a mechanical keyboard with a small Linux SBC that supports display port over USB-C so we can use it with our AR glasses. Bonus points if it has a battery.