Welcome to Jagadeesh’s Portfolio
I am V. S. Jagadeesh, a Computer Science and Engineering graduate from Amrita Vishwa Vidyapeetham, currently working as an Apprentice at Standard Chartered, specifically focusing on SAP Analytics Cloud.
During my undergraduate studies, I was selected as an Exchange Student at the University of Twente, where I worked on my Bachelor’s thesis under the supervision of Dr. Lorenzo. My work focused on building and analyzing a Graph-Augmented Retrieval-Augmented Generation (RAG) system, Weaving Stories, for generating story like narratives grounded in historical facts by connecting Large Language Models to external history datasets. I evaluated multiple RAG variants (Hybrid, Semantic, and Graph-based) through a user study, analyzing coherence, understandability, and engagement, with Graph-based RAG showing the strongest overall performance.
Through this project and my prior work in machine learning, I developed a strong interest in how large language models behave internally, particularly in scenarios involving hallucinations, unsafe outputs, and reliability failures. I am especially interested in the intersection of LLM interpretability, safety, and evaluation, with the goal of understanding why models fail and how they can be made more trustworthy in real world deployment.
Alongside my technical work, I value clear communication and structured thinking, which I have developed through leadership roles, including serving as an Area Director in Toastmasters International (District 92, 2024–25). Outside of work, I enjoy reading crime and thriller novels.
I’m always interested in discussions and collaborations around LLM interpretability, safety, factuality, evaluation, and real world AI reliability!