Greg Priday gregpriday

Lambda Labs PyTorch Training Environment

This CLAUDE.md provides context for Claude Code to manage PyTorch model training on Lambda Labs GPU infrastructure.

Important Notes

Always run Claude Code from a project directory, not from home directory
This is a single-purpose training environment - no virtual environments needed
Git LFS is often not pre-installed - check and install if needed
NumPy 2.x breaks system PyTorch - always use NumPy <2.0

GitHub Issue Analysis Guide for AI Agents

Your Role: You are the task planner for this issue. This is the planning/architectural phase—similar to what happens in /plan mode or when an agent analyzes before implementing.

Purpose: Provide a comprehensive codebase map for the issue, so the implementing agent can start coding immediately without extensive exploration.

Your Goal: Read the issue, explore the relevant parts of the codebase, then create a comment that gives the implementing agent a complete picture of:

What existing functions/classes to use
Where they're located
How they work together

	#!/bin/bash
	# Claude Code Setup Script - Universal Version
	# Works in both root and non-root environments
	# Automatically detects and handles sudo requirements
	set -e # Exit on error

	# Detect OS name and version for display
	if [ -r /etc/os-release ]; then
	. /etc/os-release
	OS_NAME="$NAME"

	{
	"permissions": {
	"allow": [
	"Bash",
	"Read",
	"Edit",
	"Write",
	"WebFetch",
	"Grep",
	"Glob",