Pytorch Lightning

If you're training neural networks with PyTorch and tired of writing the same device management and training loop boilerplate, this organizes your code into LightningModules with clear sections for training, validation, and optimizer config. The Trainer handles multi-GPU orchestration automatically, so switching from single GPU to DDP or FSDP is just a parameter change. The LightningDataModule pattern is genuinely helpful for keeping data pipelines clean and reusable. Built-in callbacks for checkpointing and early stopping save you from reimplementing common patterns. The real win is going from prototype to distributed training without rewriting your model code, though you'll still need to understand the underlying PyTorch when things break.

Install to Claude Code

npx -y skills add davila7/claude-code-templates --skill pytorch-lightning --agent claude-code