Quick background: I have a Math/CS degree and about 2 years of experience in CV/Speech/Diffusion models (mostly on the engineering side). I'm looking to pivot more into robotics now and I'm super curious about building a learning roadmap. I'd love to dive deeper into the software/AI side, especially things like Vision-Language-Action (VLA) models. Does anyone have advice or suggestions on a good roadmap for someone with my background who's relatively new to the robotics domain itself? Thanks in advance for any pointers!