Compile task signatures into optimized language model pipelines with DSPy.
npx skills add https://github.com/graniet/kheish --skill dspy-graniet
Please help me install this Agent Skill. Skill: dspy Source: https://github.com/graniet/kheish/tree/main/skills/mlops/research/dspy Command: npx skills add https://github.com/graniet/kheish --skill dspy-graniet
This Skill eliminates the tedious, error-prone process of manual prompt engineering for building complex language model systems like RAG pipelines, AI agents, and classifiers, enabling systematic, data-driven improvement of LM performance.
Use the dspy skill to build a chain-of-thought question-answering module and optimize it with your training data for better accuracy.
High-intent search queries and answers about installing and using this skill.
Automatic prompt optimization uses training data to systematically improve LM prompts and few-shot examples, eliminating manual prompt engineering overhead for complex LM pipelines.
You can build a RAG pipeline by defining retrieval and generation tasks using declarative type-safe signatures, then applying built-in teleprompters to automatically optimize the prompts using your historical training data.
Declarative LM programming supports Anthropic Claude, OpenAI, and local models via Ollama, allowing you to compile task signatures into optimized pipelines across multiple configured language model providers.
The best way to structure multi-stage AI agent workflows is using modular pipeline building, composing reusable declarative components for tool-using agents and multi-stage language model workflows requiring systematic optimization.
Yes, you need training data such as historical support tickets to automatically improve prompts and few-shot examples using built-in teleprompters for production-ready language model pipelines.
ruiyangruiyiCompile declarative DSPy program signatures into optimized prompt and few-shot pipelines.
ouakarDesign and deploy production AI/ML systems with MLOps pipelines and monitoring.
devCharuzuDesign and deploy production-grade RAG pipelines with chunking, embeddings, vector stores, and evaluation.
ovachieverDeclaratively program language models and automatically optimize prompts with DSPy.
