Building LLM Applications with DSPy

Get the eBook free when you register your print book at Manning.

Static, over-engineered prompts quickly lose effectiveness as models and data change. DSPy replaces inflexible text-based prompts with dynamic contract-based Python code, so your prompts can freely adapt and scale. In Building LLM Applications with DSPy, AI engineers Serj Smorodinsky and Brett Kennedy introduce the powerful DSPy framework and show you how it can revolutionize the way you think of prompt and context engineering. In this practical guide, you’ll learn how DSPy automatically optimizes context, evaluates prompt effectiveness, and automatically tweaks your prompts as models drift and change. As you go, you’ll get tips and techniques to maintain stable inference results as your apps and agents evolve.

Building LLM Applications with DSPy introduces DSPy best practices you can adopt to create reliable, production-ready systems through proper task definition, evaluation, and optimization. Practical to the core, this book helps you construct a full professional portfolio of AI applications, including an LLM-based classification system, a summarizer, and a RAG-based application. Youll build multi-step workflows using DSPys modular system, finally culminating in fully agentic pipelines, all without writing a single prompt by hand. A DSPy contributor, author Serj Smorodinsky speaks authoritatively about how to get the most out of this elegant tool. And, as with every Manning book, you’ll find a carefully constructed learning path, readable text, lots of helpful graphics, and our promise that the details are correct and reliable.

Whats inside

• Define prompts as Python classes
• Optimize prompts automatically for higher accuracy
• Structure complex tasks into simple modules

About the reader

For anyone working directly with LLMs, with basic Python skills.

About the author

Serj Smorodinsky is a contributor to DSPy, a data scientist, and an AI engineer with over ten years of combined experience in software development and data science. His work spans NLP for customer-service related conversational AI, agentic workflow automation, and LLM evaluation, with hands-on experience leading teams to build chatbots and retrieval-augmented systems for enterprise clients. He also teaches agentic systems and data science in production at Nebius Academy (formerly Y-Data School of Data Science).

Brett Kennedy is a data scientist with over thirty years of software development experience and more than ten in data science. He is a regular contributor to open source projects and an author of data science blog articles. He is also the author of Outlier Detection in Python.