Define and orchestrate AI workflows with declarative LM programming.
npx skills add https://github.com/erfanzar/Xerxes-Agents --skill dspy-erfanzar
Please help me install this Agent Skill. Skill: dspy Source: https://github.com/erfanzar/Xerxes-Agents/tree/main/src/python/xerxes/skills/research/dspy Command: npx skills add https://github.com/erfanzar/Xerxes-Agents --skill dspy-erfanzar
Declarative LM programming provides a structured way to define and compose AI workflows, reducing ad-hoc prompt engineering.
Install DSPy and run a basic module to see declarative LM programming in action.
High-intent search queries and answers about installing and using this skill.
Declarative LM programming structures modular AI workflows using signatures and modules, replacing ad-hoc prompt engineering with type-checked inputs and outputs. This enables safe, reusable composition for complex reasoning pipelines.
RAG pipelines are optimized using teleprompters, which automatically refine declarative LM modules. This approach treats prompting as an optimization problem rather than manual tuning, yielding production-ready retrieval-augmented generation systems.
Multi-stage reasoning models support serialization for saving and loading states. This allows you to persist optimized pipelines, enabling production deployment and consistent experimentation across different environments.
Declarative LM programming orchestrates tool-using agents by defining structured inputs and outputs via signatures. These modular pipelines compose multi-stage reasoning and tool usage, supporting both research and production settings.
Teleprompters improve self-improving pipelines by systematically optimizing declarative modules. Instead of manual prompt adjustments, teleprompters automatically tune the LM calls within your workflow to achieve better performance.
Devsoul2026Build modular AI pipelines with DSPy signatures and optimizers.
masermediagroup-stackDeploy and manage LLM and ML model-serving stacks with optimized backends.
InverterNetworkStore and retrieve conversation context using Pinecone vector databases.
huidgeDefine typed signatures and compose pipelines with the DSPy framework.
