In an old AYRS publication, I saw mention of sprung plank beams for multihulls. Cut two identical plywood planks, tapered if you want to. Install a wooden slat with a right-angled triangular section on each edge of one of them so that the right angle is on the outside of the plank, and the hypotenuse of the triangle on the inside. Lay the plank with the triangular edges down and right down the middle, install a vertical shear web out of plywood. (Would a plywood shear web be stronger if it were cut so that the wood fibres were angled at + and - 45 degrees instead of 0 degrees and 90 degrees?) That shear web needs to be just high enough that when you put the second plank on top and bend them both, they make a 90 degree corner on each side (or whatever other angle you gave that triangular slat) and you can glue the planks together along their edges. A section through the beam has a lenticular shape.
The people discussing this way of making beams liked them, but did not mention any scantling rules or equations for engineering. They were building small boats, and perhaps those were cheap enough that they could afford applying an experimental approach.
I would check whether the wood's modulus is similar enough to glass that it makes sense to let them share the load, or whether glass is only good to keep a slightly thicker layer of epoxy on the wood to protect it better against water. I find that epoxy without glass eventually cracks. Once I noticed in time, once not. I read that WEST System epoxy supposedly has a maximum elongation (is it 3%?) that is less than the maximum movement of wood as it takes on and sheds water vapour that still goes through the epoxy, and that is what makes it crack. In my experience, a thin layer of glass helps, though if I ever use epoxy on plywood on a structure that is supposed to last, I would look for a less brittle resin.
Then again, there are vastly more experienced people than me here who may correct me.
