N SANJAY RAO

I’m an AI/GenAI developer with 2 years of hands-on experience evaluating LLM outputs on a Google GenAI content-evaluation workflow at Cognizant. I verify 150+ cases daily at 99%+ accuracy against evolving policy guidelines, and I’ve progressed through eight increasingly technical evaluation workflows.

I build document Q&A experiences end-to-end, including an end-to-end RAG chatbot that parses PDF/DOCX/CSV, uses hybrid FAISS + BM25 retrieval, applies cross-encoder reranking, and streams LLM responses. I designed numbered citations back to source chunks, plus query expansion and caching to improve quality and efficiency—running on free tiers.

My focus is strong Python-based engineering for classical ML/NLP and production-ready LLM systems. I mentor L1 analysts on evaluation accuracy and consistency, and I’m seeking AI engineering roles where I can build knowledge-base assistants, document Q&A pipelines, and internal automation.