What problem does it solve? Manual prompt engineering is fragile, hard to maintain, and difficult to improve systematically. DSPy replaces hand-tuned prompts with declarative signatures and modules, then uses data-driven optimizers to automatically improve prompts and few-shot examples. ## Core Features & Use Cases - Declarative Signatures & Modules: Define tasks as input-output signatures and compose them with Predict, ChainOfThought, ReAct, and ProgramOfThought modules. - Automatic Optimization: Use BootstrapFewShot, MIPRO, and BootstrapFinetune to generate few-shot demonstrations and better instructions from training data. - Complex Pipelines: Build multi-stage RAG systems, tool-using agents, classifiers, and structured extraction with Pydantic-typed outputs. - Use Case: Build a multi-hop RAG question-answering system, then optimize it with BootstrapFewShot on 50 labeled examples to measurably improve answer accuracy without rewriting prompts. ## Quick Start Use the dspy skill to build a chain-of-thought question answering module and optimize it with BootstrapFewShot on my training examples.
