I'm also interested. I would focus on making relatively minor changes to AIF over time, while developing a new installation system in parallel. The new installation system could be made available for testing early on, to gather feedback, but should not replace the old AIF until it was good enough (for some definition of good enough).
If you don't think this is a horrible plan and, after having discussed with me here, don't think I'm a horrible person, then I'm willing to put in some effort.
Jeremiah also seem like a fine gentleman, and I wouldn't mind the slightest if he were elected.