If DSPy is so great, why isn't anyone using it?
skylarbpayne.com
March 23, 2026
6 min read
🔥🔥🔥🔥🔥
59/100
Summary
DSPy promises to address significant challenges in AI engineering, leading to quicker model testing and improved system maintainability for users. Despite these benefits, adoption remains low among companies.
Key Takeaways
- Companies using DSPy report benefits such as faster model testing and improved system maintainability.
- The primary barrier to DSPy adoption is its complexity, which requires users to adopt unfamiliar abstractions.
- Many teams end up creating their own suboptimal versions of DSPy due to the difficulty of implementing its principles correctly.
- The evolution of AI systems often leads teams through stages of increasing complexity, from simple implementations to more structured and resilient systems.
Community Sentiment
Positives
- DSPy offers an integrated perspective that could streamline the development of AI agents, potentially improving workflow efficiency.
- The emphasis on separating prompts from code encourages a more structured approach to prompt engineering, which can enhance clarity and maintainability.
Concerns
- The community feels that LLMs are often overused, with simpler ML approaches like entity recognition being overlooked, which could lead to unnecessary latency and costs.
- There is skepticism about DSPy's focus on prompt optimization, as many existing libraries already provide similar functionalities without the added complexity.
- Concerns were raised about the article's commercial intent rather than providing valuable insights into DSPy, leading to disappointment among readers.
