{"href":"https://api.simplecast.com/oembed?url=https%3A%2F%2Fpytorch-dev-podcast.simplecast.com%2Fepisodes%2Fbackend-extensibility-966RCMHR","width":444,"version":"1.0","type":"rich","title":"Backend extensibility","thumbnail_width":300,"thumbnail_url":"https://image.simplecastcdn.com/images/8cefde76-fb46-406a-8d87-ab0df67f3423/92f11400-2dad-49b4-8b14-cce35f5ab765/pytorch-symbol-02-orangeondark.jpg","thumbnail_height":300,"provider_url":"https://simplecast.com","provider_name":"Simplecast","html":"<iframe src=\"https://player.simplecast.com/56047b49-e69d-44de-8d7d-cf6a8d8bd510\" height=\"200\" width=\"100%\" title=\"Backend extensibility\" frameborder=\"0\" scrolling=\"no\"></iframe>","height":200,"description":"What's the current state of backend extensibility?  How did PyTorch evolve from being a CPU and CUDA only framework to also support AMD ROCm and XLA?  What are some problems with adding an out-of-tree backend, and what's some work to make it better?"}