Back to prompt library
DSPy Evaluation Pipeline Design
Linked challenge: Hybrid Reasoning AI Evaluation Engine
Prompt source
planning1 lines1 sectionsNo variables0 checklist items
Design a DSPy evaluation pipeline that can take an LLM's output and a corresponding benchmark question, then verify its factual accuracy using an external knowledge source (simulated via LlamaIndex RAG). Detail how DSPy modules will orchestrate calls to Gemini 2.5 Pro for verification and incorporate hybrid reasoning.