Research

I have conducted research in machine learning optimization, feature selection, privacy, and applied AI, resulting in multiple peer-reviewed publications across IEEE, Springer, and Elsevier journals, amongst others.

Current Research

I am currently a Research Assistant at the ARiSE (Advanced Research in Software Engineering) Lab at Columbia University. I am working on an NSF-funded project focused on compliance auditing of LLMs using causal inference and explainable AI techniques. The primary goal is to check adherence to FIPPs' purpose limitation and data minimization guidelines. Apart from Columbia University, there are academic collaborators from Wesleyan University and University of South Carolina, while industry participation is being spearheaded by Google and IBM whose focus is to provide LLMs and datasets.

Previous Research

• Published 9 peer-reviewed research papers with 400+ citations in areas including feature selection, stock market prediction, signature verification, social networks advertising and image segmentation.
• Collaborated with academic researchers and guided undergraduate students at CMATER (Centre for Microprocessor Applications Training and Research) Lab, Jadavpur University.
• Presented research findings at multiple international conferences organized by IEEE and Elsevier.

Tech: Python, NumPy, SciPy, scikit-learn, PyPI, MATLAB