Dhruvang Rajnikant Patel

Improved output quality for 3 enterprise NLP pipelines by annotating 10,000+ text samples using Python, PyTorch, and TensorFlow to identify false outputs and edge cases. Boosted retrieval accuracy by 40% on large-scale RAG systems by comparing vector embeddings and flagging FAISS inconsistencies. Enhanced 5 conversational AI models by writing structured commentary to guide fine-tuning. Collaborated cross-functionally to integrate agent-based AI solutions into production environments.