63-data-engineering-fine-tuning

npx skills add majiayu000/claude-skill-registry --skill 63-data-engineering-fine-tuning --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

# Build Your Data Engineering Skill Before learning data engineering for fine-tuning—preparing training datasets that transform generic models into domain specialists—you'll **own** a data engineering skill. This isn't downloading someone else's notes. You're building expertise from official documentation, creating a reusable asset that will guide you through the entire chapter and beyond. --- ## Why Skill-First? Traditional learning: Read about data engineering, then maybe create a skill later. **Skill-First learning**: Create your skill NOW, then use the chapter to test and improve it. The difference matters because: 1. **You learn faster** when you have a knowledge base to consult 2. **You retain more** when you actively improve what you've built 3. **You own an asset** at the end—not just knowledge, but reusable intelligence By Part 8, you've seen this pattern repeatedly. This chapter is no different: create the skill first, then battle-test it through real data engineering work. --- ## Step 1: Get a Fresh Skills Lab Start clean. Don't reuse old state—clone fresh: 1. Go to [github.com/panaversity/claude-code-skills-lab](https://github.com/panaversity/claude-code-skills-lab) 2.

cd claude-code-skills-lab
claude
ls -la .claude/skills/llmops-data-engineer/
cat .claude/skills/llmops-data-engineer/SKILL.md