At Eminence Technology India, I develop and deploy AI solutions that streamline internal workflows. I maintain data pipelines for near-real-time features and refine RAG workflows and LLM-based agents, using RAGAS-style evaluation to catch quality regressions.
At Asmadiya TechnologiesIndia, I architected a production RAG chatbot with LangChain and Hugging Face Transformers, evaluating it with the full RAGAS framework. I also shipped AWS Bedrock Agents for live e-commerce product search, cart management, and order placement.
I deployed models across text, vision, and speech using Triton Inference Server, TensorRT-LLM, and vLLM. I also automated a crawl-to-vector-store pipeline using ChromaDB, FAISS, and Pinecone to feed enterprise documentation into RAG systems.
