I haven’t done anything more here so far, and likely won’t for the foreseeable future. So if someone else wants to pick this up and drive it that’d be great. The only important thing here, I think, is that we pick one single solution, and that it’s a solution that we think has a good chance of surviving for a good while.
With the state of AI now, this does feel like a much more doable project today than just 6 months back.