[MLIR] Advice for newcomer

Hi there,

Currently I'm an LLVM newcomer with 1 patch accepted and committed. Also, I'm helping a research that constantly involves Deep Learning models created using TensorFlow. I'm highly intersted on starting to contribute to MLIR also, but trying to benefit that research, if possible. Perhaps contributing to optimization passes. Would you please give me some advice on where to start?, and what documentation should I start studying to accomplish it?.

Thank you.
Hamilton.