dspy

Automate end-to-end AI system building with declarative LM programming in Python.

Updated Apr 19, 2026

Search Tags:#agents #declarative #prompt-optimization #dspy #rags #lm-programming

Ready to run

npx skills add https://github.com/gqf2008/hermez-ai --skill dspy-gqf2008

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dspy
Source: https://github.com/gqf2008/hermez-ai/tree/main/skills/mlops/research/dspy
Command: npx skills add https://github.com/gqf2008/hermez-ai --skill dspy-gqf2008

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Declaratively compose AI tasks to replace brittle, hand-tuned prompts, enabling scalable and maintainable LLM pipelines.

Core Features & Use Cases

Build multi-stage AI systems (RAG, agents, classifiers) with modular components and clear data flow.
Optimize prompts automatically via DSPy optimizers, improving outputs with data.
Reuse and compose signatures and modules to accelerate development in research and production.

Quick Start

Create a simple DSPy module and run a basic QA task to see the declarative workflow in action.

Frequently Asked Questions about dspy

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I optimize LLM prompts automatically instead of manual tuning?▼

You can optimize LLM prompts automatically by using DSPy optimizers, which iteratively improve prompt outputs based on your training data to replace brittle, hand-tuned prompts.

What is declarative LM programming for building AI systems?▼

Declarative LM programming automates building end-to-end AI systems by composing modular components with clear data flow, replacing manual prompt engineering with scalable, maintainable pipelines.

Can I build multi-stage RAG pipelines and agents without writing custom prompts?▼

Yes, you can build complex multi-stage RAG pipelines and agents by reusing and composing DSPy signatures and modules, accelerating development for both research and production environments.

Do I need Python and the DSPy library to configure and run multi-stage AI pipelines?▼

Yes, you need Python and the DSPy library to configure and run these end-to-end AI pipelines, along with optional language model providers to execute the underlying tasks.

How do I replace brittle LLM pipelines with maintainable modular components?▼

Replace brittle LLM pipelines by declaratively composing AI tasks into modular components with clear data flow, enabling scalable and maintainable LLM pipelines across different providers.