Implement RAG & DSPy Optimization

Implement the RAG pipeline using LlamaIndex for a sample patent corpus (provided or create a small synthetic one). Integrate DSPy to optimize the prompts for your 'Patent Researcher' agent when performing prior art searches and for the 'Infringement Analyst' when comparing claims. Show how you've tuned the LLM calls for improved accuracy and reduced hallucination with Claude Opus 4.1.