Matmul on GPU/TPU by hand ✍️
I drew 91 frames to show how to divide large matrices into "tiles" to use accelerators such as GPU/TPU. As AI engineering jobs are getting more competitive, I believe learning PyTorch is no longer adequate. If you want to stand out, learn JAX. It gives you low-level control and runs fast on TPUs and GPUs. JAX automatically helps you tile matrix multiplications like this. It speeds up core calculations and makes you more valuable!
AI by Hand ✍️ Workshops 7/23
🔥 New: Agent (A2A), Transformer (Chinese), SOTA (XAttention)
Register (50% off): https://lu.ma/oonu4c7n?coupon=XW9R0Y-NEWSLETTER