TVM 2022 Machine Learning Compilation Course Video Lectures Triton Triton With PyTorch MLIR torch-mlir XLA AITemplate https://locust.io/ Other GitHub - mirage-project/mirage: A multi-level tensor algebra superoptimizer Generating Fast GPU Kernels without Programming in CUDA/Triton | by Zhihao Jia | Sep, 2024 | Medium Further Reading Why are ML Compilers so Hard? A friendly introduction to machine learning compilers and optimizers DL Compiler Mini-Summit at PyTorch Conference 2024 - YouTube