I went down a similar rabbit hole a while back and ended up building something that attacks the same problem from the opposite direction: instead of just a hardware box that switches everything (video + USB), I wrote Fence[1], a software layer that automates switching USB input devices when you glide the mouse to the edge of the screen.
The idea is basically "Synergy's convenience, but with real hardware switching." You run a tiny client on each machine. When your cursor hits the left edge of your desktop, Fence tells a USB and HDMI Switch to physically redirect your keyboard and mouse to the next PC.
The switching happens in hardware and you can design your layout per-direction and per-device.
Where the L1T KVM is the "one box handles video and IO beautifully" approach, Fence is more of an "IO routing layer" that lets you keep your existing monitors and their auto-input-switching (or a separate video path).
I built it specifically to be cross-platform. You don't pass clicks/keystrokes over the network, just a "switch to pc2 PC, left edge" message.
Not a replacement for the L1T if you want one-button video+peripheral switching, but if someone likes their monitor's own input handling and just wants the "mouse to edge" workflow it's a nice middle ground.
I like the fact that moving the mouse to different edges of the screen can show exactly the source to the sink that I want.
I originally built it for live streaming with OBS, but now, I miss it when I have more than one computer I need to deal with at a time.