using-pytorch-engineering

# Using PyTorch Engineering

## Overview

This meta-skill routes you to the right PyTorch specialist based on symptoms. PyTorch engineering problems fall into distinct categories that require specialized knowledge. Load this skill when you encounter PyTorch-specific issues but aren't sure which specialized skill to use.

**Core Principle**: Different PyTorch problems require different specialists. Match symptoms to the appropriate specialist skill. Don't guess at solutions—route to the expert.

## When to Use

Load this skill when:
- Working with PyTorch and encountering problems
- User mentions: "PyTorch", "torch", "CUDA", "GPU", "distributed training"
- Need to implement PyTorch models or optimize performance
- Debugging PyTorch training issues
- Setting up production PyTorch infrastructure

**Don't use for**: Framework-agnostic ML theory, non-PyTorch frameworks, algorithm selection (use training-optimization or other packs)

---

## How to Access Reference Sheets

**IMPORTANT**: All reference sheets are located in the SAME DIRECTORY as this SKILL.md file.

When this skill is loaded from:
  `skills/using-pytorch-engineering/SKILL.md`

Reference sheets like `tensor-operations-and-memory.md` are at:
  `skills/using-pytorch-engineering/tensor-operations-and-memory.md`

NOT at:
  `skills/tensor-operations-and-memory.md` ← WRONG PATH

When you see a link like `[tensor-operations-and-memory.md](tensor-operations-and-memory.md)`, read the file from the same directory as this SKILL.md.

---

## Routing by Symptom

### Memory Issues

**Symptoms**:
- "CUDA out of memory"
- "OOM error"
- "RuntimeError: CUDA out of memory"
- "GPU memory usage too high"
- "tensor memory leak"
- "memory consumption increasing"

**Route to**: See [tensor-operations-and-memory.md](tensor-operations-and-memory.md) for memory management and optimization.

**Why**: Memory management is foundational. Must understand tensor lifecycles, efficient operations, and profiling before other optimizations.

*