dspy

Compile declarative LM programs into modular, reusable pipelines.

Updated Apr 10, 2026

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

Ready to run

npx skills add https://github.com/overviewlabs/WHOX --skill dspy-overviewlabs

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

DSPy provides a framework to design, compose, and optimize complex language-model pipelines using declarative programming, reducing boilerplate and enabling self-improving AI systems.

Core Features & Use Cases

Structured, typed task definitions via Signatures for reliable LM calls and better tooling.
Multi-stage pipelines (retrieve, reason, generate) and tool-enabled agents for robust information work.
Built-in teleprompters (optimizers) like BootstrapFewShot, MIPRO, and COPRO to automatically improve prompts and training data.
Rich references and examples to accelerate adoption in research, development, and production.

Quick Start

Install DSPy, define a simple signature, and run a basic prediction to verify end-to-end functionality.

Frequently Asked Questions about dspy

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

FAQPage Schema

How do I build self-improving AI pipelines for multi-stage workflows?▼

Build self-improving AI pipelines by using declarative LM programming to compile modular, reusable stages like retrieve, reason, and generate into robust workflows. This framework applies typed signatures and built-in teleprompters to automatically optimize prompts and training data.

What is the best way to automate prompt optimization for language models?▼

Automate prompt optimization by applying built-in teleprompters like BootstrapFewShot, MIPRO, and COPRO to your declarative LM pipelines. These optimizers automatically compile and improve prompts, reducing manual boilerplate and enabling self-improving AI systems.

How do I create reliable language model calls for RAG pipelines and tool-using agents?▼

Create reliable language model calls for RAG pipelines and tool-using agents by defining structured, typed task definitions via Signatures. This typed signature system provides better tooling and ensures robust information work across multi-stage workflows.

Does declarative LM programming work for both research and production environments?▼

Declarative LM programming works for both research and production environments where teams design complex AI systems. The framework provides modular components, typed signatures, and rich references to accelerate adoption and satisfy developer needs.

When do I need to use teleprompters instead of manual prompt engineering?▼

Use teleprompters instead of manual prompt engineering when you need to automatically improve prompts and training data across multi-stage workflows. They compile declarative LM programming into modular pipelines, eliminating boilerplate and enabling self-improving AI systems.