EVAL BASICS — CHAPTERS
Chapter#	Chapter Title
Intro	Seventy Years of Software Engineering, and What Agentic AI Actually Changes
1	How LLMs Work, and Why They're Not Consistent
2	What Is an Agent, and How Do You Judge a Good One?
3	Building a Golden Dataset (Worked Example)
4	Evals vs. Classic QA — and Building a Scorecard a Team Actually Trusts
5	Goodhart's Law — When Your Eval Stops Measuring What You Think It Measures
6	The Ten Agent Fault Lines — and How Many Times You Actually Need to Run an Eval
7	Applying the Framework — Using LangSmith to Measure and Improve a Coding Agent
8	A Training Plan for Learning LangSmith
9	Knowledge Check — 25 Questions Covering Chapters 1–8
10	A Four-Agent Claude Code Pipeline for Woody's Hardware — Built and Evaluated with LangSmith