---
title: "Sayanti Bardhan - Staff Engineer in Bengaluru | Tal"
description: "Meet Sayanti Bardhan, Staff Engineer @Samsung R&D || MS by Research (CS) @ IIT-Madras || 10+ Publications based in Bengaluru. AI stack includes antigravity and opencode."
image: "https://tal.club/meet/sayanti-bardhan/share-image"
---

# Sayanti Bardhan on Tal

## Headline

Staff Engineer @Samsung R&D \|\| MS by Research \(CS\) @ IIT\-Madras \|\| 10\+ Publications

## About

• Creative and adaptable thinker with over 13\+ years of cross\-industry experience in Machine Learning, Artificial Intelligence, Deep Learning and Computer Vision\. • Over 11 publications in several conference and journals \(likes of CVPR\-W etc\.\) of repute in field of Computer Vision, Machine Learning and Deep Learning\. • Areas of Expertise: Data Science, Computer Vision, NLP, Machine Learning, Deep Learning, Artificial Intelligence, ML Model Deployment, ML/DL model Optimization, XR/AR/VR, LLM, Agentic AI, RAG • Sayanti is expert in Python, Pytorch, Keras, TensorFlow, Caffe, AWS\., C\+\+, Tableau, Power BI, Matlab, R

## Location

Bengaluru

## AI stack

antigravity

opencode

## Work experience

Staff Engineer at Samsung R&D Institute India, Jul 2023 \- Present · 3 yrs 2 mos, India, • Object Detection for Samsung XR/VR Device \- Computer Vision: – Developed and deployed a critical software component of object detection in XR device\. Architected a deep learning model with quantisation aware training, to balance high model performance with efficient model compression Visible output: Solution deployed in recently launched Samsung XR device, contributing to product’s successful go\-to\-market strategy\. Achievements: Recipient of Samsung excellence award for contribution in Samsung VST device market launch efforts\. • Fine Grain Image Classification for Mobile application \- Computer Vision: – Designed and developed novel architecture & training strategy, to obtain high accuracy and low latency mobile deep learning models\. Deployed trained models in production environments\. Visible output: Solution deployed in production of Samsung Galaxy mobile phones \- July 2024\. • LLM driven evaluation for Samsung XR device • Advanced Memory System for Multimodal Agentic AI Samsung XR device

Data Scientist \- Advanced analytics at Novartis, Jun 2021 \- Jul 2023 · 2 yrs 2 mos, • Few\-shot document classification \- NLP: – Conceptualized and designed Deep Learning based OCR and Document classification model on challenging real\-world dataset\. Savings on model deployment\- $ 20 Mn\. • AI based automatic keyword extraction \-NLP : – Developed keyword based search engine, that can serve as labels for multi\-class document classification, with limited data\. • Achievements: Recipient of 1 Star award & 2 Galaxy team award\.

Scientist C at Indian Institute of Tropical Meteorology, Jul 2016 \- Jun 2021 · 5 yrs, Pune, Maharashtra, India, • Deep Learning based climate prediction model: – Conceptualized and developed DL based global climate prediction model\. Developed model emulated the climate parameters with 90\-95 % accuracy and accelerated the computations by 8\-10 times\. • Machine Learning model to predict factors influencing evolution / intensification of a cyclone: –Modelling innovation include computation of Linear Discriminants and their projection onto Principle Component space to select the core set of variables used to predict cyclone intensity change\. –Developed ML and statistical tool based model to predict intensity changes in cyclones with 92% accuracy\. • Visible output: 9 Conference Publications and 2 Journal Publications of repute\. • Relevant Skills: Computer Vision, Object Detection, Real\-Time Processing, Deep Learning, Optimization, Machine Learning\.

Scientist at Abdus Salam International Centre for Theoretical Physics \(ICTP\), Aug 2012 \- Aug 2012 · 1 mo, Trieste Area, Italy, • Predictive Modelling for current and future climate prediction: Modelling to improve quality of monsoon forecast in India using machine learning and statistical tools\. Innovation in modelling helped to understand dependence of Indian Monsoon on ENSO mode\.

Scientist B & Trainee Scientist at Indian Institute of Tropical Meteorology, Aug 2011 \- Jun 2016 · 4 yrs 11 mos, Pune, • Signal processing for compression, filtration, detection and imaging of echo signals: –Proposed and developed signal processing algorithms for imaging echo signals with 0\.3 m resolution –Led signal processing and ML team of 4 members for first of its kind Made in India sonar system that images targets buried in seabed\. –System currently deployed and used for civilian purpose like shipwreck detection near ports etc\. • DL based Object Detection in challenging images like cluttered background, multiple objects, complex scenes etc\.: – Designed and developed real\-time detection framework for objects buried in seabed\. Achievements: This research contributed in identification and discussion of potential use cases worth $ 1\.2 Mn\.

Summer Intern at Indian Institute of Technology, Kharagpur, May 2009 \- Jun 2009 · 2 mos, Kharagpur, \- Developed a mathematical model for optimizing the complexity of the copolymerization reactor\. Furthermore, a hierarchical approach was adopted that facilitated independent testing and maintanance of each level of control\. \- Developed program to automate control required in a process plant\. This would eliminate the requirement of large scale manual operation of the controllers\.

## Education

M\.S\. \(by Research\), Computer Vision , Computer Science, Indian Institute of Technology, Madras

Bachelor of Technology \(B\.Tech\.\), Applied Electronics and Instrumentation, West Bengal University of Technology, Kolkata

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

```json
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Person",
      "@id": "https://tal.club/meet/sayanti-bardhan#person",
      "name": "Sayanti Bardhan",
      "url": "https://tal.club/meet/sayanti-bardhan",
      "image": "https://tal.club/meet/sayanti-bardhan/profile-photo.jpg",
      "description": "Meet Sayanti Bardhan, Staff Engineer @Samsung R&D || MS by Research (CS) @ IIT-Madras || 10+ Publications based in Bengaluru. AI stack includes antigravity and opencode.",
      "jobTitle": "Staff Engineer",
      "worksFor": {
        "@type": "Organization",
        "name": "Samsung R&D Institute India"
      },
      "hasOccupation": {
        "@type": "Occupation",
        "name": "Staff Engineer"
      },
      "homeLocation": {
        "@type": "Place",
        "name": "Bengaluru"
      },
      "alumniOf": [
        {
          "@type": "EducationalOrganization",
          "name": "Indian Institute of Technology, Madras"
        },
        {
          "@type": "EducationalOrganization",
          "name": "West Bengal University of Technology, Kolkata"
        }
      ],
      "knowsAbout": [
        "antigravity",
        "opencode"
      ],
      "mainEntityOfPage": {
        "@id": "https://tal.club/meet/sayanti-bardhan#webpage"
      }
    },
    {
      "@type": "ProfilePage",
      "@id": "https://tal.club/meet/sayanti-bardhan#webpage",
      "url": "https://tal.club/meet/sayanti-bardhan",
      "name": "Sayanti Bardhan on Tal",
      "description": "Meet Sayanti Bardhan, Staff Engineer @Samsung R&D || MS by Research (CS) @ IIT-Madras || 10+ Publications based in Bengaluru. AI stack includes antigravity and opencode.",
      "dateModified": "2026-08-01T20:26:53.586Z",
      "isPartOf": {
        "@id": "https://tal.club/#website"
      },
      "mainEntity": {
        "@id": "https://tal.club/meet/sayanti-bardhan#person"
      },
      "primaryImageOfPage": {
        "@type": "ImageObject",
        "url": "https://tal.club/meet/sayanti-bardhan/profile-photo.jpg",
        "contentUrl": "https://tal.club/meet/sayanti-bardhan/profile-photo.jpg",
        "caption": "Sayanti Bardhan profile photo on Tal"
      },
      "inLanguage": "en-IN"
    }
  ]
}
```