Profile

I am a researcher in AI and Deep Learning with several years of academic and industrial research experience and a strong focus on Computer Vision, Multimodal Learning, Explainable AI, and Generative AI. My expertise includes designing and creating novel deep learning architectures, as well as optimizing models like CNNs, RNNs, GANs, Transformers and diffusion models according to tasks and domains of interest. I am proficient with cutting-edge models such as ResNet, Mask R-CNN, YOLO, Transformers, DDPM, DDIM, and have utilized advanced techniques, including data augmentation, transfer learning, hyperparameter optimization, and ensemble learning to enhance model performance. I have experience in interdisciplinary collaboration, ensuring effective teamwork on highly technical and novel projects.

My current role as a Senior Engineer I, R&D, at Vasoptic Medical, Inc., USA, involves research in computer vision, deep learning, medical image analysis, explainable AI, with applications in Laser Speckle Contrast Imaging (LSCI) of the eye. At present, I am based in Maryland, USA. Prior to this, I worked as a Data Analytics Consultant with the same company. I was a Project Guide at Institute of Human Resources Development (IHRD), Kerala, India, before joining Vasoptic Medical Inc. I have experience as a Software Engineer in R&D at Vehere, where I was supervised by Dr. Kankar Shubra Dasgupta. I have worked as a Junior Research Fellow at Department of Computer Science and Engineering, Indian Institute of Technology Kharagpur, under the guidance of Prof. Pabitra Mitra with funding from Indian Space Research Organisation (ISRO).

I received my B.Tech. (Honours) degree in Electronics & Electrical Communication Engineering from Indian Institute of Technology Kharagpur (Kharagpur, WB, India) with a Minor in Physics, and my Master of Science in Engineering (MSE) degree in Electrical & Computer Engineering from Johns Hopkins University (Baltimore, MD, USA).

During my Master's at JHU, I worked on "Development of efficient, autonomous asteroid detection and tracking algorithms on satellite time-series images, for real-time onboard applications on spacecraft hardware", funded by NASA, in collaboration with JHU Applied Physics Laboratory. At Vehere my major research project was conceptualizing and designing algorithms for Drone (Unmanned Aerial Vehicle) Detection and Identification for real-time commercial application. At IHRD, I supervised several research teams on Deep Learning and Computer Vision R&D projects using Python (TensorFlow, Keras, PyTorch, OpenCV), C++, MATLAB. My goal is to use AI and Machine Learning technology to develop and improve human lives for the benefit of society.

Research Interests

• Multimodal Learning
• Computer Vision
• Deep Learning
• Explainable AI (XAI)
• Generative AI
• Physics-informed machine learning (PIML)
• Natural Language Processing