At Tech Mahindra, I worked with Python, Generative AI, LLM concepts, and AI application development during my AI Engineer internship. I also developed AI workflows using RAG, embeddings, vector search, FastAPI, and REST APIs.
I built an AI Document Assistant that retrieves relevant document chunks and supplies them as context to an LLM for document-based answers. I refined chunking and retrieval settings and validated generated answers against retrieved context.
For Real-Time User Intent Recovery AI, I developed an ML-based approach to analyze user-behaviour data and predict purchase intent. I addressed data quality challenges through validation, preprocessing, feature engineering, and model evaluation.
I also designed Buddy, an AI productivity assistant that interprets voice or text commands for file-management operations. I worked on safeguards including command validation, path verification, permission checks, and confirmation before sensitive actions, where implemented.
