Skills & Expertise

Led the integration of Model Context Protocol (MCP) into the Agents product, allowing ingestion of any data source from MCP servers. Designed Agent–MCP interoperability for seamless contextual augmentation and multi-source reasoning.

Designed a mechanism to extract keywords from user queries and map them to relevant tags using Named Entity Recognition. Matched extracted tags with database table columns, leveraging Redis for efficient lookup and similarity scoring.

Engineered a highly efficient chatbot pipeline for ML research papers, utilizing OpenAI LLM. Curated and processed a dataset of 100 papers and implemented advanced text-cleaning and token-based segmentation for rapid answer retrieval in Flask. Pioneered a dynamic question-to-answer pipeline with context extraction using cosine similarity and cluster prediction with an accuracy of 89%.