{"href":"https://api.simplecast.com/oembed?url=https%3A%2F%2Fpytorch-dev-podcast.simplecast.com%2Fepisodes%2Fjust-enough-cuda-to-be-dangerous-4J1DrjLZ","width":444,"version":"1.0","type":"rich","title":"Just enough CUDA to be dangerous","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/9ee6bfc2-07d2-48e4-89d1-16f80205a57f\" height=\"200\" width=\"100%\" title=\"Just enough CUDA to be dangerous\" frameborder=\"0\" scrolling=\"no\"></iframe>","height":200,"description":"Ever wanted to learn about CUDA but not sure where to start? In this sixteen minute episode I try to jam in as much CUDA knowledge as could be reasonably expected in a podcast. You won't know how to write a kernel after this episode, but you'll know about what a GPU is, what the general CUDA programming model is, why asynchronous execution makes everything complicated, and some general principles PyTorch abides by when designing CUDA kernels."}