Rajarshi Dutta profile photo 1 on Tal

Rajarshi Dutta

American Express | UofT | IITK

Gurugram

4+ years of work experience

American Express
Feb 20251 yr 4 mosPresent

Gurugram, Haryana, India

Commercial marketing modelling - GDS Working on marketing models via Email and Direct Mail channel and implementing a unified propensity model for entire US customer base for commercial card products. • Received “Top Contributor Award - Q3’25”
Decision Point Analytics
Aug 20246 mosJan 2025

Gurugram, Haryana, India

• Developed RAG-based insight generation tool for client sales reports using LangChain, Pinecone VectorDB, and chain-of-thought reasoning, enabling automated query-driven analysis and reducing manual report review time • Streamlined SKU-level promotion modeling across 500+ SKUs in 12 markets using discount adjustments, VARMAX, OLS models • Integrated an end-to-end scoring pipeline for monthly inference and periodic model refresh for Bimbo Chile and Spain clients
University of Toronto
May 20233 mosJul 2023

Toronto, Ontario, Canada

Supervisor: Oleksandr Voznyy Clean Energy Lab, UTSC • Implemented message passing using relational ConvNet incorporating bidirectional updates and residual connections, CGCNN for band gap and orbital energy predictions for bilinear zincblende and rocksalt crystal structures • Created a custom magpie, orbital-aware and structural embeddings thereby achieving an MSE score of 0.03 and 0.23 • Devised an XGBoost band-gap prediction model from 2048 sized organic Morgan fingerprints, achieving an MSE of 0.30
Aerial Robotics IITK
Jun 20221 yr 2 mosJul 2023

IIT Kanpur, India

Aerial Robotics IITK
Apr 20211 yr 5 mosAug 2022

Kanpur, Uttar Pradesh, India

An enthusiastic bunch of folks excited about Aerial vehicles. Explored sub domains like SLAM integrations, hardware components, object detection, map generation using ROS melodic. Successfully built multiple drones and participated in various competitions with podium finishes.
Astronomy Club, IIT Kanpur
May 20221 yr 2 mosJun 2023

Indian Institute of technology Kanpur

• Mentored 10+ juniors on computational astronomical techniques in simulations and data processing • Developed algorithms from scratch for exoplanet detection using transit photometry and identification and classification of solar flares from light curve datasets • Published a fully functional python package named Jux on Python Package Index (PyPi)

Rajarshi’s AI Stack

hugging face
hugging face
langchain
langchain
langgraph
langgraph
openai
openai

Education

Indian Institute of Technology, Kanpur

Bachelor of Technology, Materials science and engineering

Indian Institute of Technology, Kanpur

20204 yrs2024

NA, Schooling, PCM

St. Xavier's School, Raiganj

20182 yrs2020

Join Rajarshi and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Rajarshi Dutta on Tal

Headline

American Express | UofT | IITK

About

Data Scientist at American Express under the commercial marketing modelling team. Passionate about ML and love building things from scratch. Open for collaborating on any sort of related interesting projects.

Location

Gurugram

AI stack

hugging face

langchain

langgraph

openai

Work experience

Data Scientist at American Express, Feb 2025 - Present · 1 yr 4 mos, Gurugram, Haryana, India, Commercial marketing modelling - GDS Working on marketing models via Email and Direct Mail channel and implementing a unified propensity model for entire US customer base for commercial card products. • Received “Top Contributor Award - Q3’25”

Data Scientist at Decision Point Analytics, Aug 2024 - Jan 2025 · 6 mos, Gurugram, Haryana, India, • Developed RAG-based insight generation tool for client sales reports using LangChain, Pinecone VectorDB, and chain-of-thought reasoning, enabling automated query-driven analysis and reducing manual report review time • Streamlined SKU-level promotion modeling across 500+ SKUs in 12 markets using discount adjustments, VARMAX, OLS models • Integrated an end-to-end scoring pipeline for monthly inference and periodic model refresh for Bimbo Chile and Spain clients

Research Intern at University of Toronto, May 2023 - Jul 2023 · 3 mos, Toronto, Ontario, Canada, Supervisor: Oleksandr Voznyy Clean Energy Lab, UTSC • Implemented message passing using relational ConvNet incorporating bidirectional updates and residual connections, CGCNN for band gap and orbital energy predictions for bilinear zincblende and rocksalt crystal structures • Created a custom magpie, orbital-aware and structural embeddings thereby achieving an MSE score of 0.03 and 0.23 • Devised an XGBoost band-gap prediction model from 2048 sized organic Morgan fingerprints, achieving an MSE of 0.30

Team Head at Aerial Robotics IITK, Jun 2022 - Jul 2023 · 1 yr 2 mos, IIT Kanpur, India

Junior Member at Aerial Robotics IITK, Apr 2021 - Aug 2022 · 1 yr 5 mos, Kanpur, Uttar Pradesh, India, An enthusiastic bunch of folks excited about Aerial vehicles. Explored sub domains like SLAM integrations, hardware components, object detection, map generation using ROS melodic. Successfully built multiple drones and participated in various competitions with podium finishes.

Project Mentor at Astronomy Club, IIT Kanpur, May 2022 - Jun 2023 · 1 yr 2 mos, Indian Institute of technology Kanpur, • Mentored 10+ juniors on computational astronomical techniques in simulations and data processing • Developed algorithms from scratch for exoplanet detection using transit photometry and identification and classification of solar flares from light curve datasets • Published a fully functional python package named Jux on Python Package Index (PyPi)

Education

Bachelor of Technology, Materials science and engineering, Indian Institute of Technology, Kanpur

NA, Schooling, PCM, St. Xavier's School, Raiganj

Public links

github: https://github.com/Rajarshi1001