Engineers Are Reinventing DSPy's Wheel Without Knowing It, Experts Warn

Engineering teams are unknowingly rebuilding DSPy's core architecture from scratch, experts warn, as the powerful AI pipeline framework's 4.7M monthly downloads pale against LangChain's 222M despite offering superior structure, optimization, and composability that teams eventually recreate anyway — the hard way.

• DSPy, despite offering typed I/O, composable modules, and built-in optimization for AI pipelines, sees only 4.7M monthly downloads compared to LangChain's 222M, largely because its abstractions require engineers to think differently before they have experienced the pain of building without them.
• Engineering teams consistently reinvent DSPy's core patterns on their own, progressing through stages of prompt management, structured outputs, retries, RAG, evals, and model abstraction, ultimately building a more complex and bug-prone version of what DSPy already provides out of the box.
• Whether teams adopt DSPy directly or borrow its principles, experts are urging engineers to implement typed I/O, separated prompts, composable units, early eval infrastructure, and model abstraction from day one, warning that skipping these patterns leads to unmaintainable AI systems down the road.