Supervised fine-tuning, preference optimization, reinforcement learning, and alignment stack that turns pretrained base models into production-grade, aligned models on Trainium - the main lever on delivered model quality and the fastest-growing, least-covered workload class on Trainium today. We work end-to-end across a vertically integrated stack - hardware, compiler, runtime, PyTorch/JAX (XLA), and the open-source post-training ecosystem (TRL, veRL, OpenRLHF, NeMo-Aligner, Ray) - turning research techniques into production-grade, high-throughput systems that run efficiently across thousands of nodes.