---
title: "Devraj Mandal - Senior Data Scientist in Bengaluru | Tal"
description: "Meet Devraj Mandal, Senior AI Researcher | PhD (IISc) | Computer Vision & Multimodal GenAI | Ex-Rakuten based in Bengaluru. AI stack includes claude code, codex, and cursor."
image: "https://tal.club/meet/devraj-mandal/share-image"
---

# Devraj Mandal on Tal

## 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

[View the original profile](https://tal.club/meet/devraj-mandal)

```json
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Person",
      "@id": "https://tal.club/meet/devraj-mandal#person",
      "name": "Devraj Mandal",
      "url": "https://tal.club/meet/devraj-mandal",
      "image": "https://tal.club/meet/devraj-mandal/profile-photo.jpg",
      "description": "Meet Devraj Mandal, Senior AI Researcher | PhD (IISc) | Computer Vision & Multimodal GenAI | Ex-Rakuten based in Bengaluru. AI stack includes claude code, codex, and cursor.",
      "jobTitle": "Senior Data Scientist",
      "worksFor": {
        "@type": "Organization",
        "name": "PATHOMIQ, Inc."
      },
      "hasOccupation": {
        "@type": "Occupation",
        "name": "Senior Data Scientist"
      },
      "homeLocation": {
        "@type": "Place",
        "name": "Bengaluru"
      },
      "alumniOf": [
        {
          "@type": "EducationalOrganization",
          "name": "Indian Institute of Science (IISc)"
        },
        {
          "@type": "EducationalOrganization",
          "name": "Jadavpur University"
        }
      ],
      "knowsAbout": [
        "claude code",
        "codex",
        "cursor",
        "hugging face",
        "notion",
        "ChatGPT"
      ],
      "mainEntityOfPage": {
        "@id": "https://tal.club/meet/devraj-mandal#webpage"
      }
    },
    {
      "@type": "ProfilePage",
      "@id": "https://tal.club/meet/devraj-mandal#webpage",
      "url": "https://tal.club/meet/devraj-mandal",
      "name": "Devraj Mandal on Tal",
      "description": "Meet Devraj Mandal, Senior AI Researcher | PhD (IISc) | Computer Vision & Multimodal GenAI | Ex-Rakuten based in Bengaluru. AI stack includes claude code, codex, and cursor.",
      "dateModified": "2026-08-28T09:09:55.503Z",
      "isPartOf": {
        "@id": "https://tal.club/#website"
      },
      "mainEntity": {
        "@id": "https://tal.club/meet/devraj-mandal#person"
      },
      "primaryImageOfPage": {
        "@type": "ImageObject",
        "url": "https://tal.club/meet/devraj-mandal/profile-photo.jpg",
        "contentUrl": "https://tal.club/meet/devraj-mandal/profile-photo.jpg",
        "caption": "Devraj Mandal profile photo on Tal"
      },
      "inLanguage": "en-IN"
    }
  ]
}
```