senior-prompt-engineer

Prompt optimization refines and tests prompt variations to improve LLM response quality, accuracy, and latency. This Skill automates A/B testing different prompts, evaluates performance against production targets, and identifies the most effective patterns for your application without manual iteration.

RAG evaluation assesses retrieval-augmented generation systems by measuring retrieval accuracy, response quality, and relevance. This Skill provides automated evaluation tools that identify performance gaps and improvement areas across your RAG pipeline.

Agent orchestration involves configuring multi-step workflows where LLMs make decisions and take actions. This Skill helps you design, configure, and deploy sophisticated agentic systems with production-grade observability, security, and reliability across inference and deployment.

Yes. This Skill integrates with LangChain, LlamaIndex, and DSPy to optimize prompts and agentic patterns across these frameworks, enabling production-grade development with consistent performance monitoring and compliance checks.

Production-grade requirements include performance targets, scalability constraints, security patterns, compliance standards, observability for monitoring, and reliable development practices across code, testing, CI/CD pipelines, and deployment workflows.

Yes. This Skill covers the complete LLM lifecycle: prompt optimization for inference, evaluation of system performance, and monitoring in production to ensure consistent quality and identify optimization opportunities over time.