Likitha AM

I’m an AI/ML and full-stack developer currently pursuing a B.E. in Computer Science and Engineering, with a focus on turning data into usable, user-facing systems. I enjoy building solutions that are measurable, explainable, and practical to deploy.

In an AI/ML internship at IBM SkillBuild, I worked on thyroid disease detection—improving prediction accuracy by 13% over baseline models through preprocessing, feature engineering, and model optimization. In my current Full Stack Development internship at IIT Ropar, I’m developing a Task Management web application using the MERN stack, including task CRUD, task tracking, RESTful APIs, and frontend integration.

My projects reflect how I like to combine LLMs with strong retrieval and structured reasoning. I built a Text-to-SQL system using a RAG pipeline with Ollama (Mistral), FAISS-based semantic schema retrieval, and automated multi-table JOIN generation; and MedGraphRAG, a GraphRAG dialysis decision support system using knowledge graphs with multi-agent workflows for explainable, confidence-scored recommendations. I’ve also reached Finalist at Smart India Hackathon 2025 and published research on stressor identification and an intelligent tutoring system.