The One-Line Slowdown: Why AI Models Get Slower

Learn more about model collapse here → https://ibm.biz/~bs0HZQOH7 Your AI model runs, but the output is wrong. Antoni Viros i Martin explains how hidden synchronization and graph breaks can quietly slow AI training and reduce compiler optimizations. Learn a practical debugging mindset for improving AI model performance in PyTorch and compiled workloads. 00:00 - The Hidden Cost of Reading a Single Tensor Value 00:13 - How Torch.compile Optimizes Your Model 00:35 - Why Accelerators Work Ahead of Python 00:48 - The .item() Problem: Python Scalar Reads 01:12 - How Graph Breaks Fragment Optimized R