Fix RAG Service & Voice Interaction Issues in Python-Based Healthcare Application I need an experienced AI/ML developer to diagnose and resolve specific technical issues in a Python-based healthcare application that uses Retrieval-Augmented Generation (RAG) for content delivery and Vapi.ai for AI voice agent integration. Specific Problems to Solve: 1. RAG System Issues: - Vector search returning irrelevant clinical content chunks - Embedding mismatches causing incorrect template retrieval - Context window optimization needed for accurate responses 2. Voice Interaction Bugs: - Speech interruption detection not triggering correctly - Silence timeout thresholds causing premature disconnections - Response retry logic failing after user interruptions 3. Content Validation: - Output filtering not enforcing template structure - Responses deviating from approved clinical script formats Required Skills: - Strong Python debugging capabilities - Hands-on experience with LangChain, LlamaIndex, or similar RAG frameworks - Experience with STT/TTS features using Vapi.ai - Prompt engineering and LLM output structuring Preferred Experience: - HIPAA-compliant application development - Healthcare data systems or EHR integration What You'll Deliver: - Root cause analysis document for each bug - Fixed Python code with inline comments explaining changes - Test results proving issue resolution - Setup/deployment notes for future maintenance