Context Engineering with DSPy

Chapter 9. Real-World Use Cases

A note for Early Release readers

With Early Release ebooks, you get books in their earliest form—the author’s raw and unedited content as they write—so you can take advantage of these technologies long before the official release of these titles.

This will be the 9th chapter of the final book. The GitHub repo will be made active later on.

If you’d like to be actively involved in reviewing and commenting on this draft, please reach out to the editor at @oreilly.com.

In Chapter 8, you learned to build agents: ReAct loops, RAG pipelines, memory strategies, and Recursive Language Models that manage their own context. You’ve got the building blocks. Now let’s see what happens when you put them together to solve real problems.

This chapter is the convergence point for everything in this book. Each use case maps directly to the failure modes from Chapter 1, applies the context engineering techniques you learned there, and uses the DSPy modules, evaluation metrics, and optimizers from Chapters 2 through 8. The table in Table 9-1 shows how each section connects to the concepts you’ve already learned.

Table 9-1. Cross-reference table mapping each use case to its failure modes, context engineering techniques, agentic patterns, and key DSPy features.

Use Case	Failure Modes	Techniques	Agentic Pattern	Key DSPy Feature
9.1 Sentiment Classifier	Context Underspecification	FewShot Learning, Prompt Optimization	Classification	ChainOfThought, BootstrapFewShotWithRandomSearch ...