The Future of Prompt Engineering: Prompts to Programs

Gradient Descent - Podcast about AI and Data

Apr 29, 2025•36 min•Season 1Ep. 4

--:--

Listen in podcast apps:

Apple Podcasts

Spotify

Download

Listen to this episode in Metacast mobile app

Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more

Episode description

Explore the evolution of prompt engineering in this episode of Gradient Descent. Manual prompt tuning — slow, brittle, and hard to scale — is giving way to DSPy, a framework that turns LLM prompting into a structured, programmable, and optimizable process.

Learn how DSPy’s modular approach — with Signatures, Modules, and Optimizers — enables LLMs to tackle complex tasks like multi-hop reasoning and math problem solving, achieving accuracy comparable to much larger models. We also dive into real-world examples, optimization strategies, and why the future of prompting looks a lot more like programming.

Listen to our podcast on these platforms:

• YouTube: https://youtube.com/@WisecubeAI/podcasts

• Apple Podcasts: https://apple.co/4kPMxZf

• Spotify: https://open.spotify.com/show/1nG58pwg2Dv6oAhCTzab55

• Amazon Music: https://bit.ly/4izpdO2

Mentioned Materials:

• DSPy Paper - https://arxiv.org/abs/2310.03714

• DSPy official site - https://dspy.ai/

• DSPy GitHub - https://github.com/stanfordnlp/dspy

• LLM abstractions guide - https://www.twosigma.com/articles/a-guide-to-large-language-model-abstractions/

Our solutions:

- https://askpythia.ai/ - LLM Hallucination Detection Tool