Headline

Senior AI Researcher | PhD (IISc) | Computer Vision & Multimodal GenAI | Ex-Rakuten

About

I am a PhD-trained AI/ML scientist and Senior Data Scientist working at the intersection of computer vision, deep learning, medical AI, and computational pathology. My work focuses on developing robust AI/ML solutions for medical image analysis, with an emphasis on scalable computer vision pipelines for whole-slide imaging and related diagnostic workflows. Over the last several years, I have built AI systems for cancer pathology and biomedical imaging, including automated biomarker analysis, nuclei segmentation, image registration and stitching for whole-slide images, molecular prediction from histopathology images, immune-cell phenotype analysis, and quantitative image-analysis pipelines. My work spans both research innovation and practical deployment, including integration with tools such as QuPath. I completed my PhD from the Indian Institute of Science, Bangalore, where my research focused on cross-modal retrieval, hashing, noisy-label learning, semi-supervised learning, incremental learning, and large-scale multimodal search. I have published in venues and journals including CVPR, ICIP, WACV, ACCV, TIP, TMM, PRL, EMBC, and Computers in Biology and Medicine. My core strengths include developing end-to-end AI pipelines, designing models under limited-label and noisy-label settings, translating research ideas into deployable systems, and solving real-world computer vision problems in healthcare, pathology, and multimodal AI. Selected highlights PhD in AI/ML from the Indian Institute of Science, Bangalore. Published research in leading computer vision, machine learning, medical imaging, and biomedical AI venues. Built AI/ML systems across computational pathology, whole-slide image analysis, medical imaging, multimodal learning, and computer vision. Recipient of multiple research and innovation awards, including Rakuten Grand Prize / innovation recognitions and academic travel grants. Key areas: Computational Pathology, Medical AI, Whole Slide Imaging, Nuclei Segmentation, WSI Classification, Multiple Instance Learning, Foundation Models, Deep Learning, Computer Vision, Multimodal Learning, PyTorch, TensorFlow, QuPath.

Location

Bengaluru

AI stack

claude code

codex

cursor

hugging face

notion

ChatGPT

Work experience

Senior Data Scientist at PATHOMIQ, Inc., Aug 2025 - Present · 1 yr 1 mo, California, United States, *Building AI/ML systems for computational pathology and medical image analysis, with a focus on scalable whole-slide image analysis, robust deep learning model development, and validation across diverse datasets. * Developing deep learning pipelines for large-scale biomedical image classification, segmentation, and image analysis tasks. * Working with computer vision, representation learning, and foundation-model-based approaches for pathology image understanding. * Improving model robustness and generalization across diverse datasets using augmentation, normalization, regularization, and validation strategies. * Exploring methods for aggregating local image-level predictions into higher-level slide-level insights. * Contributing to end-to-end model development, including data processing, model training, evaluation, error analysis, and performance optimization. * Collaborating on research and development efforts aimed at translating AI models into practical computational pathology workflows.

Senior Research Scientist at Rakuten Institute of Technology, Apr 2024 - Aug 2025 · 1 yr 5 mos, Bengaluru · Hybrid, Worked on applied AI, computer vision, and generative modeling systems for visual content understanding and generation. * Developed AI/ML approaches for automated layout generation involving text and image placement in visual designs. * Explored constrained optimization and diffusion-based modeling approaches for structured visual generation. * Worked on scalable vector graphics generation using computer vision and machine learning techniques. * Built LLM-assisted product annotation workflows for automated product understanding and enrichment. * Contributed to research and prototyping efforts across computer vision, generative AI, and multimodal content understanding.

Associate Research Scientist at Rakuten Institute of Technology, Apr 2020 - Apr 2024 · 4 yrs 1 mo, Bengaluru · Hybrid, Worked on AI/ML systems for computational pathology, biomedical imaging, medical image analysis, and digital health applications. * Built AI/ML pipelines for computational pathology, biomedical imaging, and quantitative medical image analysis. * Developed automated biomarker-analysis workflows on multiplexed immunofluorescence images, contributing to published work in Computers in Biology and Medicine. * Designed blur-robust nuclei segmentation models for microscopy images, contributing to published work at EMBC. * Developed whole-slide image registration and stitching workflows with practical integration into pathology analysis tools. * Worked on spatial-context-aware modeling approaches for predicting molecular signals from histopathology images. * Contributed to tumor microenvironment analysis, immune-cell phenotype characterization, and cellular-level quantitative image-analysis pipelines. * Collaborated with multidisciplinary teams across AI research, pathology, oncology, and biomedical imaging.

Research Intern at Mercedes-Benz Research and Development India, Apr 2018 - Oct 2018 · 7 mos, Bengaluru, Worked on computer vision and machine learning algorithms for action recognition in a zero-shot learning setting. * Explored methods for recognizing unseen action categories using semantic representations and transfer-learning-based approaches. * Contributed to research and prototyping efforts in the automotive AI and infotainment domain. * Applied deep learning, representation learning, and visual recognition techniques to real-world video understanding problems.

PHD Student at Indian Institute of Science (IISc), Aug 2014 - Mar 2020 · 5 yrs 8 mos, Bangalore, Conducted doctoral research in machine learning, computer vision, and multimodal representation learning. * Developed algorithms for cross-modal retrieval, hashing, semi-supervised learning, noisy-label learning, incremental learning, and large-scale visual search. * Worked on learning robust representations under challenging real-world constraints such as limited labels, noisy supervision, unpaired data, multi-label settings, and scalable retrieval requirements. * Published research in leading computer vision and machine learning venues including CVPR, ICIP, ACCV, WACV, TIP, TMM, and PRL. * Collaborated with academic researchers on problems involving multimodal learning, image retrieval, zero-shot learning, novelty detection, and representation learning. * Mentored junior researchers and students, supported research discussions, and contributed as a teaching assistant for optimization coursework.

Programmer Analyst Trainee at Cognizant, Jul 2011 - Apr 2012 · 10 mos, Kolkata Area, India

Education

Doctor of Philosophy (Ph.D.), Electrical Engineering Technologies/Technicians, Indian Institute of Science (IISc)

Master's Degree, Electrical Engineering Technologies/Technicians, Jadavpur University