Performance Engineering on Neuron - How to Optimize Your LLM with NKI

Learn to maximize the performance of large language models (LLMs) on AWS purpose-built accelerators through profiling techniques and custom kernel development using the Neuron Kernel Interface (NKI). Discover how to analyze what happens inside language models during inference, identify performance bottlenecks, and build optimized custom kernels to achieve superior performance. Explore practical approaches to performance engineering that can significantly speed up LLM inference workloads on AWS Neuron-powered infrastructure.