Vikhram S is an undergraduate student in Electronics and Communication Engineering at Saveetha Engineering College, India. His work focuses on machine learning systems, with particular emphasis on vision–language models, explainable AI, and domain-specific intelligence systems.
His research explores the integration of perception and reasoning in multimodal architectures. He has developed systems that combine vision transformers with language models for structured report generation in clinical settings, incorporating interpretability through attention mechanisms and gradient-based methods. His work also examines reliability in AI systems through confidence-aware inference and structured outputs.
He has worked on domain-specific AI systems including applications in healthcare and legal intelligence. His projects include a vision–language system for radiological report generation and a retrieval-based natural language processing system for querying constitutional documents. These systems emphasize reproducibility, modular design, and real-world deployment.
His work has been presented at the India AI Impact Summit (MeitY + UN Women), where he contributed to research on gender-responsive AI systems. He is also a contributor to applied AI initiatives that explore the intersection of technology, policy, and societal impact.
In addition to research, he has been involved in mentoring and academic activities as an advisor to a machine learning student community, where he has conducted workshops on AI systems and guided student projects focused on end-to-end system development.
He is currently pursuing his undergraduate degree at Saveetha Engineering College, with a focus on developing scalable and interpretable AI systems for scientific and real-world applications.