I'm Prateek: CUDA Developer, Deep Learning Engineer and a lover of high performance computing
I love writing compilers in MLIR and learning how internals of Inference engines and Deep Learning Libraries work
I am currently working on writing a compiler for distributed training and inference which takes the best of all the inference engines and deep learning libraries